DETAILED ACTION
Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
The following is a Non-Final Office Action in response to communications received November 14, 2025.. Claim(s) 3-5 and 11-20 have been canceled. Claim 1 has been amended. New claims 21-33 have been added. Therefore, claims 1-2, 6-10 and 21-33 are pending and addressed below.
Continued Examination Under 37 CFR 1.114
A request for continued examination under 37 CFR 1.114, including the fee set forth in 37 CFR 1.17 (e), was filed in this application after final rejection. Since this application is eligible for continued examination under 37 CFR 1.114, and the fee set forth in 37 CFR 1.17(e) has been timely paid, the finality of the previous Office action has been withdrawn pursuant to 37 CFR 1.114. Applicant’s submission has been entered.
Priority
Application No. 17/717,802 filing and priority data is 04/11/2022
Assignee/Applicant: AT&T Intellectual Property I, L.P.
Inventors: Zhou, Zhengyi, Guo, Jing, Hsu, Wen-Ling, Zavesky, Eric
Response to Amendment/Arguments
Claim Rejections - 35 USC § 101
Applicant's arguments filed November 14, 2025 have been fully considered but they are not persuasive.
In the remarks applicant argues that the claimed subject matter provides technical improvements to efficiency and accuracy of automated negotiation system through computational techniques. Applicant points to Recentive Analytics v Fox decision relied upon in the previous Office action which deemed generic applications of machine learning, which the current claims distinguish themselves from. Specifically, the limitations address deficiencies in current negotiation methods through application of reinforcement learning, with adaptations of proposals and computational constraints governed by defined business rules. The combined features produce a system that optimizes negotiation strategies while addressing computational inefficiencies and adapting to user-specific input not achievable through conventional/generic automation. Applicant’s argument is not persuasive. The limitations directed toward determining outcomes which include purchase agreements, the generation of expected outcomes based on analysis of interaction history data, user state and possible actions according to business rules or the performing of simulation for a business process with an expected outcome using a second model. The subsequent steps for selecting actions and updating the data , user state and model and determination of whether the outcome has been obtained and if not refining the possible actions are not processes directed toward any underlying technology but rather the abstract idea. Applicant has not explained what limitations address computational inefficiencies or even the root of the computational inefficiencies is in the technology or technical process. The rejection is maintained.
In the remarks applicant argues that the previous office action determination that the claimed technical operations are recited at a high level lacking technical details mischaracterizes the claimed limitations. Applicant points to the USPTO 2025 memo with respect to the evaluation of claim limitations as a whole for step 2A and 2B. Applicant argues the involving of machine learning in the claims demonstrate improvement to computer functionality or solving technical problems, specifically accomplish goals by implementation of RL technology in negotiations. The claimed RL model is trained on user historical interaction data of a user population segment having payment risk level similar to user engaging in negotiations in a training process. The process of prioritizing recency and training based on groups of users of similar profiles and provide RL models that can predict and adapt negotiation proposals. Applicant’s argument is not persuasive. Applicant’s arguments are directed toward using RL technology rather than improving technology without any specific technical process that goes beyond high level function to provide an expected result. The rejection is maintained.
In the remarks applicant argues the limitations introduce iterative refinement mechanism refining action spaces, proposals and expected outcomes, based on updated histories and user responses. The iterative feedback loop enables personalization optimization of negotiation outcomes for users enhancing accuracy, efficiency and adaptability. These features integrate the judicial exception into a practical application. Applicant’s argument is not persuasive. The courts have held that iterative calculations as claimed merely apply technology to analyze data and apply feedback data and other data for use in providing expected outcomes. The rejection is maintained
In the remarks applicant argues the claimed limitations introduce technical limitations tied to business rules that guide and enhance system functionality. The claims of Recentive lack analogous restraints as set forth in the claims. The claimed subject matter employs narrowing of action spaces, e.g. evaluating expected outcomes based on shortest computational distance to a desirable negotiated goal and applying termination conditions such as reaching iteration limits, user termination or predefined business rule criteria. The claimed subject matter safeguard against unnecessary computational cycles creating boundaries for optimized negotiations. Therefore, the subject matter improves computer systems functions and enforce meaningful limits on the process for managing resources or optimizing data systems under 2B. Applicant’s argument is not persuasive. In the Recentive, the courts found that the limitations and specification did not support that the focus of the application of ML models was machine learning technology, instead it was focused on the abstract idea being performed using ML technology. The current application similarly apply the RL model for the same purpose. With respect to the argument “narrowing of action spaces, e.g. evaluating expected outcomes based on shortest computational distance” improves technology or computer functionality, the specification does not support the argument. The specification ¶ 0036 discloses “the simulation process is performed iteratively at each step of the interaction, to generate an expected outcome more closely approximating the desirable outcome (i.e., minimizing a distance in the action space between the expected outcome and the desirable outcome). In additional embodiments, a simulation process can be performed before the interaction (using the most recent user state available), with the proposal/action accessed at runtime.”, The process “generate an expected outcome more closely approximating the desirable outcome” is merely the generation of a result of an analysis and does not impact, improve or change the capacity or functionality of any recited underlying technology or provide a solution to a problem rooted in technology. The rejection is maintained.
In the remarks applicant argues the refinement of action spaces as claimed strengthens computational efficiency by reducing the scope of potential outcomes at each stage of the negotiations. Applicant argues that each iteration produces optimized proposals based on feedback and constrained by business rules refining plausible action. The refinement provide increase efficiency for matching user preferences to proposals enabling successful agreements, payment plans or purchases, while reducing computational overhead. The iterative process transforms and improves automated negotiation methods that do not respond to varying user profiles and inputs. Applicant argues the leveraging RL techniques and enforcing strict constrains provides improvements to negotiation modelling and implementation. Applicant argues that the process represent a practical and transformative application of machine learning in solving technological problems in automated negotiations which is distinctive from the mere application of ML technology to perform the abstract idea as found in Recentive decision. Applicant argues the limitations do not simply apply generic ML algorithms to automate routine processes, but instead recite tailored implementation of RL training, user segmentation enhancing predictive outcomes and proposal refinements that adapt to user specific preferences and interactions. The limitations improve performance, accuracy and adaptability of negotiation systems solving critical inefficiencies in conventional methods and therefore is patent eligible. Applicant’s argument that the limitations are directed toward improving technology with a process that represent a practical and transformative application of machine learning in solving technological problems in automated negotiations which is distinctive from the mere application of ML technology to perform the abstract idea as found in Recentive decision, is not persuasive. Applicant’s own arguments makes clear that the improvement is in the abstract idea and not technology itself. The examiner disagrees that that the claimed application of the RL model for use in data analysis, generation of outcomes, refinement of results by performing iterative processing using feedback data and user interaction data which is not directed toward improving ML technology or any other technology of the claim limitations or technical field. The RL model as claimed is merely acting as a tool to perform the abstract idea used for analyzing inputted data and generating a result based on the data generated and the simulation of outcomes of a business process. The claimed technology is explicitly analogous to Recentive v Fox. The rejection is maintained.
Claim Rejections - 35 USC § 103
The amendments submitted November 14, 2025 are sufficient to overcome the prior art rejection set forth in the previous Office Action. The examiner withdraws the 103 rejection of claims.
Claim Rejections - 35 USC § 101
35 U.S.C. 101 reads as follows:
Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.
Claims 1-2, 6-10 and 21-33 are rejected under 35 U.S.C. § 101 because the instant application is directed to non-patentable subject matter. Specifically, the claims are directed toward at least one judicial exception without reciting additional elements that amount to significantly more than the judicial exception. The rationale for this determination is in accordance with the guidelines of PTO, applies to all statutory categories, and is explained in detail below.
In reference to Claims 1-2, 6-10 and 23-25:
STEP 1. Per Step 1 of the two-step analysis, the claims are determined to include a method, as in independent Claim 1 and the dependent claims. Such methods fall under the statutory category of "process." Therefore, the claims are directed to a statutory eligibility category.
STEP 2A Prong 1. The claimed invention is directed to an abstract idea without significantly more. Method claim 1 recites steps 1) analyzing data comprising user profile 2) determining desirable outcome of interaction between agent and equipment 3) training a model 4) determining desired outcome of user interaction and agent 5) determining a plurality of actions according to business rules 6) deriving a secondary model (7) performing simulation of interaction and possible actions (8) selecting next action based on comparison of outcomes (9) communicating a proposal (10) receiving a response to selected action (11) updating data, user state and (12) determining whether desirable outcome obtained (13) refining plausible action space/outcome (14) determining user behavior (15) selecting new action path (16) performing simulation of a further interaction by generating second expected outcomes for a sample of plurality of possible actions (17)selecting a further action from plurality of possible actions (18)communicating second proposal. The claimed limitations which under its broadest reasonable interpretation, covers performance of analyzing data for a business process.
When considered as a whole the claimed subject matter in light of the specification, the subject matter is directed toward analyzing data; determining outcome; constructing models; determining user state; performing simulation of next step to generate next step; selecting the next step; receiving response; updating data, state, model; determining whether desirable outcome has been obtained; refining possible actions; determining user behavior in real time and selecting new action path for user, refining actions of possible actions to a reduced number of possible actions, performing simulation of a further step of the interactions by generating second expected outcomes, selecting a further action and communication of proposals/results is directed toward determining opportunities for deeper and more personalized interactions in a business environment. The specification discloses the “personalizing of the first model” a personalized interaction with a user to obtain a desirable outcome (para 0013) where the system delivers services, resolved customer issues provides remote learning with personalized treatment (para 0026). This allows the system to use a procedure to a personalized action path based on user historical and profile data (para 0028). The specification discloses the first/primary model is used as a reference/starting model from which a secondary model can be derived using additional optimization/domain strategies. The secondary model optimized/personalized for a singular user, the users state or expected experience results from decisions which makes clear that the focus of the invention as a whole is directed toward commercial interactions. Such concepts can be found in the abstract category of commercial interactions and marketing. These concepts are enumerated in Section I of the 2019 revised patent subject matter eligibility guidance published in the federal register (84 FR 50) on January 7, 2019) is directed toward abstract category of methods of organizing human activity.
STEP 2A Prong 2: The identified judicial exception is not integrated into a practical application because the claims fail to provide indications of patent eligible subject matter that integrate the alleged abstract idea into a practical application. The additional elements recited in the claim beyond the abstract idea include a processing system comprising a processor, a primary reinforcement learning (RL) model and a secondary reinforcement learning (RL) model.
The primary reinforcement learning model does not perform any of the recited steps of the method.
The processing system applied to perform the method steps “communicating …proposal”, “receiving …a response” and “communicating ….a second proposal”, which according to MPEP 2106.05(d) II (see also MPEP 2106.05(g)) the courts have recognized the following computer functions are claimed in a merely generic manner (e.g., at a high level of generality) where technology is merely applied to perform the abstract idea or as insignificant extra-solution activity.
Receiving or transmitting data over a network, e.g., using the Internet to gather data, Symantec, 838 F.3d at 1321, 120 USPQ2d at 1362 (utilizing an intermediary computer to forward information); TLI Communications LLC v. AV Auto. LLC, 823 F.3d 607, 610, 118 USPQ2d 1744, 1745 (Fed. Cir. 2016) (using a telephone for image transmission); OIP Techs., Inc., v. Amazon.com, Inc., 788 F.3d 1359, 1363, 115 USPQ2d 1090, 1093 (Fed. Cir. 2015) (sending messages over a network); buySAFE, Inc. v. Google, Inc., 765 F.3d 1350, 1355, 112 USPQ2d 1093, 1096 (Fed. Cir. 2014) (computer receives and sends information over a network); but see DDR Holdings, LLC v. Hotels.com, L.P., 773 F.3d 1245, 1258, 113 USPQ2d 1097, 1106 (Fed. Cir. 2014)
The additional element “processing system” applied to perform the steps “analyzing …data…”, “determining…desirable outcome…”, “determining…a user state…”, “determining …a plurality of possible actions…”, “updating …data,…user state and…model”, “determining…whether desirable outcome obtained” and “determining… behavior of the user” where the steps performed by the processing system used for data analysis recited at a high level without any technical details applied to provide expected outcomes.
The additional element “processing system” applied to perform the steps “selecting…a next action…”, “selecting…a next action…” and “selecting…”further action”, where the steps performed by the processing system used for selecting business related actions recited at a high level without any technical details applied to provide expected outcomes.
The additional element “processing system” applied to perform the operations “training …a primary reinforcement learning (RL) model…”, “deriving….a secondary RL model from the primary RL model”, where the steps performed by the processing system used for training models, the limitations and specification do not provide any details as to how the training is implemented and therefore is so broad as to include any known means using generic programming. The specification discloses ¶ 0034 “the RL model can be trained to apply weighting factors (e.g. give greater weight to more recent interaction data). In another embodiment, models for a population, a population segment, and/or an individual user can be combined to construct a hybrid model”, ¶ 0039 “procedure 203 for training a reinforcement learning model and performing run-time simulations of proposed actions in an interaction with a user, in accordance with embodiments of the disclosure. The RL model is trained 231 using user features and proposals/actions to predict an outcome ( e.g. a response from the user). In an embodiment, the model is trained to map the most recent user state and a proposed action to an outcome for that action.”; ¶ 0045 …”The system then builds and trains a reinforcement learning (RL) model to map the user state and a system action (e.g. a proposal to be presented to the user) to an expected outcome (step 2408). This model may be trained using data from an individual user, a segment of similar users, or a population of users….” and wherein ¶ 0098 describes training data which can be acted upon. However the specification is silent with respect the technical details or process for “training” RL models focusing only on the function and outcome. The additional element “processing system” applied to perform the step “performing….a simulation of a next step…” and “performing…a simulation of a further step…” where the steps performed by the processing system used for simulating steps in a business activity recited at a high level without any technical details applied to provide expected outcomes.
The specification discloses such simulations as generating expected outcomes for each of a plurality of possible actions which is mere automation of data analysis by repeatedly analyzing each possible action - .(MPEP 2106.05(d) II (ii) Bancorp Services v. Sun Life, 687 F.3d 1266, 1278, 103 USPQ2d 1425, 1433 (Fed. Cir. 2012) ("The computer required by some of Bancorp’s claims is employed only for its most basic function, the performance of repetitive calculations, and as such does not impose meaningful limits on the scope of those claims.
The functions are directed toward performing analysis on human behavior and commercial activity. The additional elements further include a primary and secondary model which similarly being applied to analyze data, perform simulations and output results. The claim limitations use the model for data analysis and provides no details as to technical implementation. The limitations recite the training limitation so broad as to encompass simple programming or any and all means for training models to perform expected activities. The limitations do not delineate how the “training”, the :”use” and “deriving” of the models achieves an improvement to technology or part of a solution to a problem rooted in existing technology. The only thing the claims discloses is about the use of the model in a commercial environment The functions are is recited at a high-level of generality such that it amounts to no more than applying the exception using generic computer components. Taking the claim elements separately, the operation performed by the Method at each step of the process is purely in terms of results desired and devoid of implementation of details. This is true with respect to the limitations “training a model” and the modelling of the data, as the claimed limitations do not provide any technology to perform the recited functions. Technology is not integral to the process as the claimed subject matter is so high level that any generic programming could be applied and the functions could be performed by any known means. Furthermore, the claimed functions do not provide an operation that could be considered as sufficient to provide a technological implementation or application of/or improvement to this concept (i.e. integrated into a practical application).
When the claims are taken as a whole, as an ordered combination, the combination of limitations 1-2 and 3 are analyzing data and determining outcomes of interaction applied in the training of a model – a common business practice-and abstract construct- see Alice. The wherein clause does not further limit the determining step but rather limits outcomes to be evaluated which is user preference for communication means, determination of user preference is not a process directed toward technology. The combination of limitations 1-3 and 4-12 are directed toward deriving a second model from first model for modelling human behavior, updating data, user state and the model to determine an outcome which is directed toward data analysis and applying models for the analysis- an abstract idea – see Alice. The combination of limitations 1-12 and 13-15 is directed toward determining and refining the result of the analysis of limitations 1-12 and directed toward gathering addition user behavior data where the result are applied to select a new action path for the user- directed toward a business practice. The combination of limitations 1-16 merely apply the processing system. The combination of limitations 1-15 and 16-18 is directed toward performing additional simulations and selecting further actions for possible actions with a desirable outcome selected based on minimizing distance between expected outcomes and desirable outcome and then communicating the result – which is merely applying the processor to analyze business data and output the result. The combinations of parts is not directed toward any technical process or technological technique or technological solution to a problem rooted in technology. Accordingly, when the claims are taken as a whole, as an ordered combination, the combination of steps are not integrate the judicial exception into a practical application as the claim process fails to impose meaningful limits upon the abstract idea. This is because the claimed subject matter as a whole is directed modelling human behavior and refining the simulations- a process directed toward analyzing human activity. Therefore, the claimed limitations fail to provide additional elements or combination or elements to apply or use the judicial exception in a manner that imposes a meaningful limit on the judicial exception. The functions recited in the claims recite the concept of analyzing and determining data for analysis in a model, selecting actions based on simulation outcomes and updating data, user state and model to refine possible action which is a process directed toward analysis of human behavior and commercial activity where technology is merely being applied to implement the abstract idea.
Similar to the claim limitations, the specification lacks technical disclosure of the “training” of the primary or the “deriving” of the secondary model, instead only states that the primary model is trained and the secondary model can be derived using optimization/domain strategies. The specification discloses the “domain” to be business/rules preferences or domain expertise as constraints on the action space and/or input in selecting the next action. The domain experts can design a set of desirable action paths and the system can determine at each step of the interaction the best path for the user (para 0029). The specification discloses that the model is trained to map user state and a proposal to an expected outcome, where the model is “trained” on user population, on a segment of user population on an individual or combination of user and interaction type which focuses on the data applied for analysis rather than technical disclosure (para 0033, par 0045). The specification discloses the training of the model as performing simulation of proposed actions in an interaction with a user using user features, proposals/actions to predict an outcome (para 0039) which is an application of the model and not technical details, again lacking technical disclosure. The specification further states the system simulating possible next steps in the interaction to predict outcomes of different proposals/outcomes in the action space using the trained model and then the system choosing the action/proposal that results in best expected outcome (para 0046). The specification discloses employing classifiers explicitly trained (generic training data) as well as explicitly trained (observing user behavior, operator preferences, historical data and extrinsic data) where the classifier is used to determine predetermined criteria which benefit maximum number of subscribers.
The integration of elements do not improve upon technology or improve upon computer functionality or capability in how computers carry out one of their basic functions. The integration of elements do not provide a process that allows computers to perform functions that previously could not be performed. The integration of elements do not provide a process which applies a relationship to apply a new way of using an application. The instant application, therefore, still appears only to implement the abstract idea to the particular technological environments apply what generic computer functionality in the related arts. The steps are still a combination made to automated workflow to model a business process and does not provide any of the determined indications of patent eligibility set forth in the 2019 USPTO 101 guidance. The additional steps only add to those abstract ideas using generic functions, and the claims do not show improved ways of, for example, a particular technical function for performing the abstract idea that imposes meaningful limits upon the abstract idea. Moreover, Examiner was not able to identify any specific technological processes that goes beyond merely confining the abstract idea in a particular technological environment, which, when considered in the ordered combination with the other steps, could have transformed the nature of the abstract idea previously identified. Accordingly, this additional element does not integrate the abstract idea into a practical application because it does not impose any meaningful limits on practicing the abstract idea. The claim is directed to an abstract idea.
STEP 2B; The claim(s) does/do not include additional elements that are sufficient to amount to significantly more than the judicial exception because as discussed above with respect to concepts of the abstract idea into a practical application. The additional elements recited in the claim beyond the abstract idea include a processing system comprising a processor, a primary reinforcement learning (RL) model and a secondary reinforcement learning (RL) model.-–is purely functional and generic. Nearly every computer system will include a “processor” capable of performing the basic computer functions of “analyzing”, “determining”, ‘training”, “deriving”, “performing simulation”, “selecting”, “communicating”, “receiving”, “updating”, “modifying” and “refining” steps required by the method claims. The claimed primary reinforcement learning (RL) model does not perform any of the limitations.
Taking the claim elements separately, the function performed by the computer at each step of the process is purely conventional. Using a processing system processor to train and derive models, and using such models to determine results of a user action or plurality of actions, perform simulations and refining possible actions of behavior ----are some of the most basic functions of a computer.
When the claims are taken as a whole, as an ordered combination, the combination of steps does not add “significantly more” by virtue of considering the steps as a whole, as an ordered combination. All of these computer functions are generic, routine, conventional computer activities that are performed only for their conventional uses. See Elec. Power Grp. v. Alstom S.A., 830 F.3d 1350, 1353 (Fed. Cir. 2016). Also see In re Katz Interactive Call Processing Patent Litigation, 639 F.3d 1303, 1316 (Fed. Cir. 2011) Absent a possible narrower construction of the terms “analyzing”, “determining”, “training”, “deriving”, “performing simulation”, “selecting”, “receiving”, “updating”, “refining” and “selecting”.. are functions can be achieved by any general purpose computer without special programming. None of these activities are used in some unconventional manner nor do any produce some unexpected result. Applicants do not contend they invented any of these activities. In short, each step does no more than require a generic computer to perform generic computer functions.
As to the data operated upon, "even if a process of collecting and analyzing information is 'limited to particular content' or a particular 'source,' that limitation does not make the collection and analysis other than abstract." SAP America, Inc. v. Invest Pic LLC, 898 F.3d 1161, 1168 (Fed. Cir. 2018). Considered as an ordered combination, the computer components of Applicant’s claimed functions add nothing that is not already present when the steps are considered separately. The sequence of data reception-analysis modification-transmission is equally generic and conventional. See Ultramercial, Inc. v. Hulu, LLC, 772 F.3d 709, 715 (Fed. Cir. 2014) (sequence of receiving, selecting, offering for exchange, display, allowing access, and receiving payment recited as an abstraction), Inventor Holdings, LLC v. Bed Bath & Beyond, Inc., 876 F.3d 1372, 1378 (Fed. Cir. 2017) (sequence of data retrieval, analysis, modification, generation, display, and transmission), Two-Way Media Ltd. v. Comcast Cable Communications, LLC, 874 F.3d 1329, 1339 (Fed. Cir. 2017) (sequence of processing, routing, controlling, and monitoring). The ordering of the steps is merely being implemented by generic computer devices. The analysis concludes that the claims do not provide an inventive concept because the additional elements recited in the claims do not provide significantly more than the recited judicial exception.
According to 2106.05 well-understood and routine processes to perform the abstract idea is not sufficient to transform the claim into patent eligibility. As evidence the examiner provides:
With respect to application of model- see Alice; In re Ferguson, 558 F.3d 1359, 1364, 90 USPQ2d 1035, 1039-40 (Fed. Cir. 2009).
The specification discloses:
[00064] With reference again to FIG. 4, the example environment can comprise a
computer 402, the computer 402 comprising a processing unit 404, a system memory 406
and a system bus 408. The system bus 408 couples system components including, but
not limited to, the system memory 406 to the processing unit 404. The processing unit
404 can be any of various commercially available processors. Dual microprocessors and
other multiprocessor architectures can also be employed as the processing unit 404.
[00065] The system bus 408 can be any of several types of bus structure that can
further interconnect to a memory bus (with or without a memory controller), a peripheral
bus, and a local bus using any of a variety of commercially available bus architectures.
The system memory 406 comprises ROM 410 and RAM 412. A basic input/output
system (BIOS) can be stored in a non-volatile memory such as ROM, erasable
programmable read only memory (EPROM), EEPROM, which BIOS contains the basic
routines that help to transfer information between elements within the computer 402, such as during startup. The RAM 412 can also comprise a high-speed RAM such as
static RAM for caching data.
[00093] The terms "first," "second," "third," and so forth, as used in the claims, unless
otherwise clear by context, is for clarity only and doesn't otherwise indicate or imply any
order in time. For instance, "a first determination," "a second determination," and "a third
determination," does not indicate or imply that the first determination is to be made
before the second determination, or vice versa, etc.
[000100] Further, the various embodiments can be implemented as a method, apparatus
or article of manufacture using standard programming and/or engineering techniques to
produce software, firmware, hardware or any combination thereof to control a computer
to implement the disclosed subject matter. The term "article of manufacture" as used
herein is intended to encompass a computer program accessible from any computer readable
device or computer-readable storage/communications media. For example,
computer readable storage media can include, but are not limited to, magnetic storage
devices (e.g., hard disk, floppy disk, magnetic strips), optical disks (e.g., compact disk
(CD), digital versatile disk (DVD)), smart cards, and flash memory devices (e.g., card,
stick, key drive). Of course, those skilled in the art will recognize many modifications
can be made to this configuration without departing from the scope or spirit of the
various embodiments.
[000104] As employed herein, the term "processor" can refer to substantially any
computing processing unit or device comprising, but not limited to comprising, singlecore
processors; single-processors with software multithread execution capability; multicore
processors; multi-core processors with software multithread execution capability;
multi-core processors with hardware multithread technology; parallel platforms; and
parallel platforms with distributed shared memory. Additionally, a processor can refer to
an integrated circuit, an application specific integrated circuit (ASIC), a digital signal
processor (DSP), a field programmable gate array (FPGA), a programmable logic
controller (PLC), a complex programmable logic device (CPLD), a discrete gate or
transistor logic, discrete hardware components or any combination thereof designed to perform the functions described herein. Processors can exploit nano-scale architectures
such as, but not limited to, molecular and quantum-dot based transistors, switches and
gates, in order to optimize space usage or enhance performance of user equipment. A
processor can also be implemented as a combination of computing processing units.
With respect to the limitations “training” and “deriving” model, the specification discloses
[0033]… In another embodiment, a primary RL model can be used as a starting or reference model from
which a secondary model can be derived through methods utilizing additional optimization or domain adaptation strategies. These secondary models may have the same functional capabilities as the primary RL model; alternatively, the secondary models may be further optimized or personalized for a singular user, the user's state, or
an expected experience resulting from decisions in the action space.
[0034]… “the RL model can be trained to apply weighting factors (e.g. give greater weight to more recent interaction data). In another embodiment, models for a population, a population segment, and/or an individual user can be combined to construct a hybrid model”,
[0039]… procedure 203 for training a reinforcement learning model and performing run-time simulations of proposed actions in an interaction with a user, in accordance with embodiments of the disclosure. The RL model is trained 231 using user features and proposals/actions to predict an outcome ( e.g. a response from the user). In an embodiment, the model is trained to map the most recent user state and a proposed action to an outcome for that action.”;
[0045]… “The system then builds and trains a reinforcement learning (RL) model to map the user state and a system action (e.g. a proposal to be presented to the user) to an expected outcome (step 2408). This model may be trained using data from an individual user, a segment of similar users, or a population of users….”
Paragraph 0098 describes training data which can be acted upon.
The specification discloses the “personalizing of the first model” a personalized interaction with a user to obtain a desirable outcome (para 0013) where the system delivers services, resolved customer issues provides remote learning with personalized treatment (para 0026). This allows the system to use a procedure to a personalized action path based on user historical and profile data (para 0028). The specification discloses the first/primary model is used as a reference/starting model from which a secondary model can be derived using additional optimization/domain strategies. The secondary model optimized/personalized for a singular user, the users state or expected experience results from decisions. The specification lacks technical disclosure of the “deriving” of the secondary model, instead only states that the secondary model can be derived using optimization/domain strategies. The specification discloses the “domain” to be business/rules preferences or domain expertise as constraints on the action space and/or input in selecting the next action. The domain experts can design a set of desirable action paths and the system can determine at each step of the interaction the best path for the user (para 0029). The specification discloses that the model is trained to map user state and a proposal to an expected outcome, where the model is “trained” on user population, on a segment of user population on an individual or combination of user and interaction type which focuses on the data applied for analysis rather than technical disclosure (para 0033, par 0045). The specification discloses the training of the model as performing simulation of proposed actions in an interaction with a user using user features, proposals/actions to predict an outcome (para 0039) again lacking technical disclosure. The specification further states the system simulating possible next steps in the interaction to predict outcomes of different proposals/outcomes in the action space using the trained model and then the system choosing the action/proposal that results in best expected outcome (para 0046). The specification discloses employing classifiers explicitly trained (generic training data) as well as explicitly trained (observing user behavior, operator preferences, historical data and extrinsic data) where the classifier is used to determine predetermined criteria which benefit maximum number of subscribers. The specification lacks technical details on the classifier functional process. The examiner therefore, disagrees that the use of the ML models as claimed are processes which cannot reasonably be performed using generic processes. According to Electric Power Group decision, the “selecting information, by content or source, for collection, analysis, and display does nothing significant to differentiate a process from ordinary mental processes, whose implicit exclusion from § 101 undergirds the information-based category of abstract ideas.” The courts distinguish between claims that recite mental processes performed by humans and claims that recite mental processes performed on a computer. As the Federal Circuit has explained "[c]ourts have examined claims that required the use of a computer and still found that the underlying, patent-ineligible invention could be performed via pen and paper or in a person’s mind." Versata Dev. Group v. SAP Am., Inc., 793 F.3d 1306, 1335, 115 USPQ2d 1681, 1702 (Fed. Cir. 2015). See also Intellectual Ventures I LLC v. Symantec Corp., 838 F.3d 1307, 1318, 120 USPQ2d 1353, 1360 (Fed. Cir. 2016). The use of a physical aid (e.g., pencil and paper or a slide rule) to help perform a mental step (e.g., a mathematical calculation) does not negate the mental nature of the limitation, but simply accounts for variations in memory capacity from one person to another. For instance, in CyberSource, the court determined that the step of "constructing a map of credit card numbers" was a limitation that was able to be performed "by writing down a list of credit card transactions made from a particular IP address." In making this determination, the court looked to the specification, which explained that the claimed map was nothing more than a listing of several (e.g., four) credit card transactions. The court concluded that this step was able to be performed mentally with a pen and paper, and therefore, it qualified as a mental process. 654 F.3d at 1372-73, 99 USPQ2d at 1695. See also Flook, 437 U.S. at 586, 198 USPQ at 196 (claimed "computations can be made by pencil and paper calculations"); University of Florida Research Foundation, Inc. v. General Electric Co., 916 F.3d 1363, 1367, 129 USPQ2d 1409, 1411-12 (Fed. Cir. 2019) (relying on specification’s description of the claimed analysis and manipulation of data as being performed mentally "‘using pen and paper methodologies, such as flowsheets and patient charts’") The best paths as disclosed in the specification as performed the models to be trained and derived is similar as a process to the mental process found in the court to be analogous to the mapping process or flow sheets. (MPEP 2106.04 (a) (2) section III B and C3). The claimed structures (primary/secondary models) are generic computer components and tools to perform the mental processes. The computer components are recited at a high level of generality and merely automates functions that could reasonable be performed using mental concepts, therefore acting as a generic computer to perform the abstract idea.
Additional evidence includes:
EP 3675008 A1 by Ma et al; US Pub No. 2022/0294710 A1 by Swvigaradoss et al; US RE46153 E Makagon et al; US Patent No. 7,039,166 B1 by Peterson et al; US Patent No. 6,937,705 B1 by Godfrey et al; US Patent No. 6,898,277 B1 by Meteer et al; US Patent No. 6,823,054 B1 by Suhm et al;
The instant application, therefore, still appears to only implement the abstract ideas to the particular technological environments using what is generic components and functions in the related arts. The claim is not patent eligible.
The remaining dependent claims—which impose additional limitations—also fail to claim patent-eligible subject matter because the limitations cannot be considered statutory. In reference to claims 2, 6-10 and 23-25 these dependent claim have also been reviewed with the same analysis as independent claim 1. Dependent claim 2 is directed toward data acted upon in the determination of user state and mapping user states to the action– lacks technical disclosure focusing on the conceptual idea of mapping user data and actions in a commercial environment. Dependent claim 6 is directed toward applying AI model to provide analysis on data and a corresponding output- well known and understood application of technology. Dependent claims 7 and 8 are directed toward human behavior for interaction- a common business practice. Dependent claim 9 is directed toward comparison of expected outcomes based on business criteria- a common business practice. Dependent claim 10 is directed toward interaction without desirable outcome exceeding threshold- a common business practice. Dependent claim 23 is directed toward interaction concluded without desirable outcome in accordance with user terminating interaction-directed toward human behavior. Dependent claim 24 is directed toward interaction concluded without desirable outcome in accordance with business strategy requiring interaction to conclude (end)- a business process. Dependent claim 25 is directed toward primary and secondary RL models incorporate preferences from user historical interaction data – analyzing human behavior for application of data analysis.
The dependent claim(s) have been examined individually and in combination with the preceding claims, however they do not cure the deficiencies of claim 1. Where all claims are directed to the same abstract idea, “addressing each claim of the asserted patents [is] unnecessary.” Content Extraction & Transmission LLC v. Wells Fargo Bank, Nat 7 Ass ’n, 776 F.3d 1343, 1348 (Fed. Cir. 2014). If applicant believes the dependent claims 2, 6-10 and 23-25 are directed towards patent eligible subject matter, they are invited to point out the specific limitations in the claim that are directed towards patent eligible subject matter.
In reference to Claims 21 and 26-29:
STEP 1. Per Step 1 of the two-step analysis, the claims are determined to include a device, as in independent Claim 21 and the dependent claims. Such devices fall under the statutory category of "machine." Therefore, the claims are directed to a statutory eligibility category.
STEP 2A Prong 1. The claimed invention is directed to an abstract idea without significantly more. Device claim 21 recites operational process of a) analyzing data comprising user profile b) determining desirable outcome of interaction between agent and equipment c) determining user state d) training a model e) initiating a negotiation between user and agent f) simulating plurality of proposals g) selecting best expected outcome h) communicating proposal associated with best outcome i) obtaining a response from user equipment j) updating data k) determining whether response indicates user agreed to proposal l) refining first plausible action in response to suer has not agreed to proposal m) simulating a second plurality of proposals n)selecting subsequent best expected outcome o) communicating proposal p) obtaining a response q) updating interaction history r) repeating steps l) -q) iteratively and based on distance between expected outcomes and desirable outcomes threshold terminate negotiation, fixed iterations reached conclude method or user agrees to proposals.
The claimed limitations which under its broadest reasonable interpretation, covers performance of the limitation in the mind but for the recitation of a “device” comprising a “processing system”, “memory that stores instructions” and when “executed by the processing system facilitates the operations” above, , nothing in the claim element precludes the step from practically being performed in the mind. The human mind through mental processes is capable of analyzing data, determining desirable outcomes, determining a user state, initiating a negotiation between user and agent, selecting best expected outcome, obtaining a response, updating interaction data using memory, determining whether response indicates user agreed, refining plausible action based indications user did not agree, selecting a second plurality of proposals, updating data and repeated the method process.
With respect to the limitations “training a RL model for generating outcomes”, “simulating proposals” and “updating the RL model”, a computer model is an abstract mathematical representation of a real-world event, system or behavior. A computer model is designed to behave just like real-life events, systems. In order to train a generic model, problems must be designed to allow for the model to deal with elements required, relationships/operations between these elements and pattern/rules governing these relationships. The human mind through mental processes can apply decisions to determine elements, determine relationships/operations between elements, modifying plausible action space/outcome and pattern/rules governing these relationships. Therefore, the limitations training, deriving a model and refining actions to perform simulations, mimic mental processes of planning and decision. The claimed structures (primary/secondary models) are generic computer components and tools to perform the mental processes. The computer components are recited at a high level of generality and merely automates functions that could reasonable be performed using mental concepts, therefore acting as a generic computer to perform the abstract idea. The high level generic functions recited in the claims as being performed using first and second learning the analysis finds that the operations performed by the models can reasonably be performed using mental processes. This is because the steps of analyzing data, determining desirable outcome, determining user state, performing a simulation, selecting a next action, receiving next step, updating data, user state and determining desirable outcome, determining user behavior and selecting new action path for user, mimic human thought processes of observation, evaluation and decision, where the data interpretation is perceptible only in the human mind. See In re TLI Commc'ns LLC Patent Litig., 823 F.3d 607, 611 (Fed. Cir. 2016); FairWarning IP, LLC v. Iatric Sys., Inc., 839 F.3d 1089, 1093-94 (Fed. Cir. 2016)
The claimed limitations which under its broadest reasonable interpretation, covers performance of analyzing data for a business process. When considered as a whole the claimed subject matter in light of the specification, the subject matter is directed toward analyzing data; determining outcome; training models; determining user state; performing simulations, selecting outcomes; receiving response; updating data, state, model; determining whether desirable outcome has been obtained; determining user behavior in real time, refining actions of possible actions to a reduced number of possible actions, performing simulation of a further step of the interactions by generating second expected outcomes, selecting a further action and communication of proposals/results is directed toward determining opportunities for deeper and more personalized interactions in a business environment. The specification discloses the “personalizing of the first model” a personalized interaction with a user to obtain a desirable outcome (para 0013) where the system delivers services, resolved customer issues provides remote learning with personalized treatment (para 0026). This allows the system to use a procedure to a personalized action path based on user historical and profile data (para 0028). Such concepts can be found in the abstract category of commercial interactions and marketing. These concepts are enumerated in Section I of the 2019 revised patent subject matter eligibility guidance published in the federal register (84 FR 50) on January 7, 2019) is directed toward abstract category of mental concepts and methods of organizing human activity.
STEP 2A Prong 2: The identified judicial exception is not integrated into a practical application because the claims fail to provide indications of patent eligible subject matter that integrate the alleged abstract idea into a practical application. The additional elements recited in the claim beyond the abstract idea include a device comprising a processing system including a processor; and a memory that stores executable instructions that, when executed by the processing system, facilitate performance of operations and an RL model.
The RL model does not perform any of the recited steps of the method.
The processing system applied to perform the operation “obtaining….response”, , which according to MPEP 2106.05(d) II (see also MPEP 2106.05(g)) the courts have recognized the following computer functions are claimed in a merely generic manner (e.g., at a high level of generality) where technology is merely applied to perform the abstract idea or as insignificant extra-solution activity.
Receiving or transmitting data over a network, e.g., using the Internet to gather data, Symantec, 838 F.3d at 1321, 120 USPQ2d at 1362 (utilizing an intermediary computer to forward information); TLI Communications LLC v. AV Auto. LLC, 823 F.3d 607, 610, 118 USPQ2d 1744, 1745 (Fed. Cir. 2016) (using a telephone for image transmission); OIP Techs., Inc., v. Amazon.com, Inc., 788 F.3d 1359, 1363, 115 USPQ2d 1090, 1093 (Fed. Cir. 2015) (sending messages over a network); buySAFE, Inc. v. Google, Inc., 765 F.3d 1350, 1355, 112 USPQ2d 1093, 1096 (Fed. Cir. 2014) (computer receives and sends information over a network); but see DDR Holdings, LLC v. Hotels.com, L.P., 773 F.3d 1245, 1258, 113 USPQ2d 1097, 1106 (Fed. Cir. 2014)
The additional element “processing system” applied to perform the operations “analyzing …data…”, “determining…desirable outcome…”, “determining…a user state…”, where the functions are performed by the processing system used for data analysis recited at a high level without any technical details applied to provide expected outcomes.
The additional element “processing system” applied to perform the operations “initiating ….negotiation”, where the steps performed by the processing system used for selecting business related actions recited at a high level without any technical details applied to provide expected outcomes.
The additional element “processing system” applied to perform the operations “training …a primary reinforcement learning (RL) model…”, where the operation performed by the processing system used for training models, the limitations and specification do not provide any details as to how the training is implemented and therefore is so broad as to include any known means using generic programming. The specification discloses ¶ 0034 “the RL model can be trained to apply weighting factors (e.g. give greater weight to more recent interaction data). In another embodiment, models for a population, a population segment, and/or an individual user can be combined to construct a hybrid model”, ¶ 0039 “procedure 203 for training a reinforcement learning model and performing run-time simulations of proposed actions in an interaction with a user, in accordance with embodiments of the disclosure. The RL model is trained 231 using user features and proposals/actions to predict an outcome ( e.g. a response from the user). In an embodiment, the model is trained to map the most recent user state and a proposed action to an outcome for that action.”; ¶ 0045 …”The system then builds and trains a reinforcement learning (RL) model to map the user state and a system action (e.g. a proposal to be presented to the user) to an expected outcome (step 2408). This model may be trained using data from an individual user, a segment of similar users, or a population of users….” and wherein ¶ 0098 describes training data which can be acted upon. However the specification is silent with respect the technical details or process for “training” RL models focusing only on the function and outcome. The additional element “processing system using the secondary RL model” applied to perform the step “performing….a simulation of a next step…” and “performing…a simulation of a further step…” where the steps performed by the processing system used for simulating steps in a business activity recited at a high level without any technical details applied to provide expected outcomes.
The limitations “simulating based on the RL model, a first plurality of proposals…”, “selecting a best expected outcome…”, “communicating the proposal…”, “updating the interaction history, user state, and RL model ….”, “determining whether response indicates the user has agreed …”, “refining the first plausible action…”simulating based on updated RL model …second plurality of proposals…”, “selecting a subsequent best expected outcome….”, “communicating the proposal…”, “updating the interaction history, user state, user profile and RL model…” and “repeating steps (l)-q) in an iterative manner …” are not tied to any technology and therefore can be performed by any known means including manual processes. Furthermore the limitations are not directed toward technology but rather data collection and analysis for a business practice.
The specification discloses such simulations as generating expected outcomes for each of a plurality of possible actions which is mere automation of data analysis by repeatedly analyzing each possible action - .(MPEP 2106.05(d) II (ii) Bancorp Services v. Sun Life, 687 F.3d 1266, 1278, 103 USPQ2d 1425, 1433 (Fed. Cir. 2012) ("The computer required by some of Bancorp’s claims is employed only for its most basic function, the performance of repetitive calculations, and as such does not impose meaningful limits on the scope of those claims.
The functions are directed toward performing analysis on human behavior and commercial activity. The additional elements further include a primary and secondary model which similarly being applied to analyze data, perform simulations and output results. The claim limitations use the model for data analysis and provides no details as to technical implementation. The limitations recite the training limitation so broad as to encompass simple programming or any and all means for training models to perform expected activities. The limitations do not delineate how the “training”, of the models achieves an improvement to technology or part of a solution to a problem rooted in existing technology. The only thing the claims discloses is about the use of the model in a commercial environment The functions are is recited at a high-level of generality such that it amounts to no more than applying the exception using generic computer components. Taking the claim elements separately, the operation performed by the Method at each step of the process is purely in terms of results desired and devoid of implementation of details. This is true with respect to the limitations “training a model” the modelling of the data, as the claimed limitations do not provide any technology to perform the recited functions. Technology is not integral to the process as the claimed subject matter is so high level that any generic programming could be applied and the functions could be performed by any known means. Furthermore, the claimed functions do not provide an operation that could be considered as sufficient to provide a technological implementation or application of/or improvement to this concept (i.e. integrated into a practical application).
When the claims are taken as a whole, as an ordered combination, the combination of limitations (a)-(b) and (c) are directed toward analyzing data and determining outcomes of interaction and user state which is for the purpose of a business practice. The combination of limitations (a)-(c) and (d) are directed toward training of a model for business proposal outcomes using data from limitations (a)-(c) applying technology for performing business proposal outcomes. The combination of limitations (a)-(d) and ( e) –(h) are directed toward initiating a negotiation, where proposals are simulated applying the limitations of (a)-(d) and the outcomes of the simulation are communicated for use in a business practice. The combination of limitations (i) –(o) are directed toward obtaining a response from the proposal of limitations (a)-(h) and updating the data for analysis and then determining whether response indicates user agreements and based on user agreement condition refining outcomes to conform to business rules and then simulating expected outcomes that are then selected and corresponding proposals communicated- directed toward analyzing business data for use on communicating business proposals of the analysis of limitations (i)-(n). The combination of limitations (a)-(n) and (p)-(q) is directed toward obtaining the response of the user of limitations (a)-(o) and updating the data and model used for the analysis of proposals and then repeating limitations (l)-(q) iteratively until the analysis computes a distance between expected outcomes and desirable outcomes or user terminates negotiation, or fixed number of iterations reached or business rules require process to conclude or user agreed to proposals– which is merely applying the technology to analyze business data and output the result until a condition is met/determined. The combinations of parts is not directed toward any technical process or technological technique or technological solution to a problem rooted in technology. Accordingly, when the claims are taken as a whole, as an ordered combination, the combination of steps are not integrate the judicial exception into a practical application as the claim process fails to impose meaningful limits upon the abstract idea. This is because the claimed subject matter as a whole is directed modelling human behavior and refining the simulations- a process directed toward analyzing human activity. Therefore, the claimed limitations fail to provide additional elements or combination or elements to apply or use the judicial exception in a manner that imposes a meaningful limit on the judicial exception. The functions recited in the claims recite the concept of analyzing and determining data for analysis in a model, selecting actions based on simulation outcomes and updating data, user state and model to refine possible action which is a process directed toward analysis of human behavior and commercial activity where technology is merely being applied to implement the abstract idea.
Similar to the claim limitations, the specification lacks technical disclosure of the “training” of the RL model, instead only states that the model is trained. The specification discloses that the model is trained to map user state and a proposal to an expected outcome, where the model is “trained” on user population, on a segment of user population on an individual or combination of user and interaction type which focuses on the data applied for analysis rather than technical disclosure (para 0033, par 0045). The specification discloses the training of the model as performing simulation of proposed actions in an interaction with a user using user features, proposals/actions to predict an outcome (para 0039) which is an application of the model and not technical details, again lacking technical disclosure. The specification further states the system simulating possible next steps in the interaction to predict outcomes of different proposals/outcomes in the action space using the trained model and then the system choosing the action/proposal that results in best expected outcome (para 0046). The specification discloses employing classifiers explicitly trained (generic training data) as well as explicitly trained (observing user behavior, operator preferences, historical data and extrinsic data) where the classifier is used to determine predetermined criteria which benefit maximum number of subscribers.
The integration of elements do not improve upon technology or improve upon computer functionality or capability in how computers carry out one of their basic functions. The integration of elements do not provide a process that allows computers to perform functions that previously could not be performed. The integration of elements do not provide a process which applies a relationship to apply a new way of using an application. The instant application, therefore, still appears only to implement the abstract idea to the particular technological environments apply what generic computer functionality in the related arts. The steps are still a combination made to automated workflow to model a business process and does not provide any of the determined indications of patent eligibility set forth in the 2019 USPTO 101 guidance. The additional steps only add to those abstract ideas using generic functions, and the claims do not show improved ways of, for example, a particular technical function for performing the abstract idea that imposes meaningful limits upon the abstract idea. Moreover, Examiner was not able to identify any specific technological processes that goes beyond merely confining the abstract idea in a particular technological environment, which, when considered in the ordered combination with the other steps, could have transformed the nature of the abstract idea previously identified. Accordingly, this additional element does not integrate the abstract idea into a practical application because it does not impose any meaningful limits on practicing the abstract idea. The claim is directed to an abstract idea.
STEP 2B; The claim(s) does/do not include additional elements that are sufficient to amount to significantly more than the judicial exception because as discussed above with respect to concepts of the abstract idea into a practical application. The additional elements recited in the claim beyond the abstract idea include a device comprising a processing system including a processor; and a memory that stores executable instructions that, when executed by the processing system, facilitate performance of operations and an RL model-–is purely functional and generic. Nearly every computer system will include a “processor” capable of performing the basic computer functions of “analyzing”, “determining”, ‘training”, “performing simulation”, “selecting”, “communicating”, “receiving”, “updating”, and “refining” steps required by the method claims. The claimed primary reinforcement learning (RL) model does not perform any of the limitations.
Taking the claim elements separately, the function performed by the computer at each step of the process is purely conventional. Using a processing system processor to train models, and using such models to determine results of a user action or plurality of actions, perform simulations and refining possible actions of behavior ----are some of the most basic functions of a computer.
When the claims are taken as a whole, as an ordered combination, the combination of steps does not add “significantly more” by virtue of considering the steps as a whole, as an ordered combination. All of these computer functions are generic, routine, conventional computer activities that are performed only for their conventional uses. See Elec. Power Grp. v. Alstom S.A., 830 F.3d 1350, 1353 (Fed. Cir. 2016). Also see In re Katz Interactive Call Processing Patent Litigation, 639 F.3d 1303, 1316 (Fed. Cir. 2011) Absent a possible narrower construction of the terms “analyzing”, “determining”, “training”, “performing simulation”, “selecting”, “receiving”, “updating”, “refining” and “selecting”.. are functions can be achieved by any general purpose computer without special programming. None of these activities are used in some unconventional manner nor do any produce some unexpected result. Applicants do not contend they invented any of these activities. In short, each step does no more than require a generic computer to perform generic computer functions.
As to the data operated upon, "even if a process of collecting and analyzing information is 'limited to particular content' or a particular 'source,' that limitation does not make the collection and analysis other than abstract." SAP America, Inc. v. Invest Pic LLC, 898 F.3d 1161, 1168 (Fed. Cir. 2018). Considered as an ordered combination, the computer components of Applicant’s claimed functions add nothing that is not already present when the steps are considered separately. The sequence of data reception-analysis modification-transmission is equally generic and conventional. See Ultramercial, Inc. v. Hulu, LLC, 772 F.3d 709, 715 (Fed. Cir. 2014) (sequence of receiving, selecting, offering for exchange, display, allowing access, and receiving payment recited as an abstraction), Inventor Holdings, LLC v. Bed Bath & Beyond, Inc., 876 F.3d 1372, 1378 (Fed. Cir. 2017) (sequence of data retrieval, analysis, modification, generation, display, and transmission), Two-Way Media Ltd. v. Comcast Cable Communications, LLC, 874 F.3d 1329, 1339 (Fed. Cir. 2017) (sequence of processing, routing, controlling, and monitoring). The ordering of the steps is merely being implemented by generic computer devices. The analysis concludes that the claims do not provide an inventive concept because the additional elements recited in the claims do not provide significantly more than the recited judicial exception.
According to 2106.05 well-understood and routine processes to perform the abstract idea is not sufficient to transform the claim into patent eligibility. As evidence the examiner provides:
With respect to application of model- see Alice; In re Ferguson, 558 F.3d 1359, 1364, 90 USPQ2d 1035, 1039-40 (Fed. Cir. 2009).
The specification discloses:
[00064] With reference again to FIG. 4, the example environment can comprise a
computer 402, the computer 402 comprising a processing unit 404, a system memory 406
and a system bus 408. The system bus 408 couples system components including, but
not limited to, the system memory 406 to the processing unit 404. The processing unit
404 can be any of various commercially available processors. Dual microprocessors and
other multiprocessor architectures can also be employed as the processing unit 404.
[00065] The system bus 408 can be any of several types of bus structure that can
further interconnect to a memory bus (with or without a memory controller), a peripheral
bus, and a local bus using any of a variety of commercially available bus architectures.
The system memory 406 comprises ROM 410 and RAM 412. A basic input/output
system (BIOS) can be stored in a non-volatile memory such as ROM, erasable
programmable read only memory (EPROM), EEPROM, which BIOS contains the basic
routines that help to transfer information between elements within the computer 402, such as during startup. The RAM 412 can also comprise a high-speed RAM such as
static RAM for caching data.
[000100] Further, the various embodiments can be implemented as a method, apparatus
or article of manufacture using standard programming and/or engineering techniques to
produce software, firmware, hardware or any combination thereof to control a computer
to implement the disclosed subject matter. The term "article of manufacture" as used
herein is intended to encompass a computer program accessible from any computer readable
device or computer-readable storage/communications media. For example,
computer readable storage media can include, but are not limited to, magnetic storage
devices (e.g., hard disk, floppy disk, magnetic strips), optical disks (e.g., compact disk
(CD), digital versatile disk (DVD)), smart cards, and flash memory devices (e.g., card,
stick, key drive). Of course, those skilled in the art will recognize many modifications
can be made to this configuration without departing from the scope or spirit of the
various embodiments.
[000104] As employed herein, the term "processor" can refer to substantially any
computing processing unit or device comprising, but not limited to comprising, singlecore
processors; single-processors with software multithread execution capability; multicore
processors; multi-core processors with software multithread execution capability;
multi-core processors with hardware multithread technology; parallel platforms; and
parallel platforms with distributed shared memory. Additionally, a processor can refer to
an integrated circuit, an application specific integrated circuit (ASIC), a digital signal
processor (DSP), a field programmable gate array (FPGA), a programmable logic
controller (PLC), a complex programmable logic device (CPLD), a discrete gate or
transistor logic, discrete hardware components or any combination thereof designed to perform the functions described herein. Processors can exploit nano-scale architectures
such as, but not limited to, molecular and quantum-dot based transistors, switches and
gates, in order to optimize space usage or enhance performance of user equipment. A
processor can also be implemented as a combination of computing processing units.
With respect to the limitations “training” model, the specification discloses
[0034]… “the RL model can be trained to apply weighting factors (e.g. give greater weight to more recent interaction data). In another embodiment, models for a population, a population segment, and/or an individual user can be combined to construct a hybrid model”,
[0039]… procedure 203 for training a reinforcement learning model and performing run-time simulations of proposed actions in an interaction with a user, in accordance with embodiments of the disclosure. The RL model is trained 231 using user features and proposals/actions to predict an outcome ( e.g. a response from the user). In an embodiment, the model is trained to map the most recent user state and a proposed action to an outcome for that action.”;
[0045]… “The system then builds and trains a reinforcement learning (RL) model to map the user state and a system action (e.g. a proposal to be presented to the user) to an expected outcome (step 2408). This model may be trained using data from an individual user, a segment of similar users, or a population of users….”
Paragraph 0098 describes training data which can be acted upon.
The specification discloses the “personalizing of the first model” a personalized interaction with a user to obtain a desirable outcome (para 0013) where the system delivers services, resolved customer issues provides remote learning with personalized treatment (para 0026). This allows the system to use a procedure to a personalized action path based on user historical and profile data (para 0028). The specification discloses the “domain” to be business/rules preferences or domain expertise as constraints on the action space and/or input in selecting the next action. The domain experts can design a set of desirable action paths and the system can determine at each step of the interaction the best path for the user (para 0029). The specification discloses that the model is trained to map user state and a proposal to an expected outcome, where the model is “trained” on user population, on a segment of user population on an individual or combination of user and interaction type which focuses on the data applied for analysis rather than technical disclosure (para 0033, par 0045). The specification discloses the training of the model as performing simulation of proposed actions in an interaction with a user using user features, proposals/actions to predict an outcome (para 0039) again lacking technical disclosure. The specification further states the system simulating possible next steps in the interaction to predict outcomes of different proposals/outcomes in the action space using the trained model and then the system choosing the action/proposal that results in best expected outcome (para 0046). The specification discloses employing classifiers explicitly trained (generic training data) as well as explicitly trained (observing user behavior, operator preferences, historical data and extrinsic data) where the classifier is used to determine predetermined criteria which benefit maximum number of subscribers. The specification lacks technical details on the classifier functional process. The examiner therefore, disagrees that the use of the ML models as claimed are processes which cannot reasonably be performed using generic processes. According to Electric Power Group decision, the “selecting information, by content or source, for collection, analysis, and display does nothing significant to differentiate a process from ordinary mental processes, whose implicit exclusion from § 101 undergirds the information-based category of abstract ideas.” The courts distinguish between claims that recite mental processes performed by humans and claims that recite mental processes performed on a computer. As the Federal Circuit has explained "[c]ourts have examined claims that required the use of a computer and still found that the underlying, patent-ineligible invention could be performed via pen and paper or in a person’s mind." Versata Dev. Group v. SAP Am., Inc., 793 F.3d 1306, 1335, 115 USPQ2d 1681, 1702 (Fed. Cir. 2015). See also Intellectual Ventures I LLC v. Symantec Corp., 838 F.3d 1307, 1318, 120 USPQ2d 1353, 1360 (Fed. Cir. 2016). The use of a physical aid (e.g., pencil and paper or a slide rule) to help perform a mental step (e.g., a mathematical calculation) does not negate the mental nature of the limitation, but simply accounts for variations in memory capacity from one person to another. For instance, in CyberSource, the court determined that the step of "constructing a map of credit card numbers" was a limitation that was able to be performed "by writing down a list of credit card transactions made from a particular IP address." In making this determination, the court looked to the specification, which explained that the claimed map was nothing more than a listing of several (e.g., four) credit card transactions. The court concluded that this step was able to be performed mentally with a pen and paper, and therefore, it qualified as a mental process. 654 F.3d at 1372-73, 99 USPQ2d at 1695. See also Flook, 437 U.S. at 586, 198 USPQ at 196 (claimed "computations can be made by pencil and paper calculations"); University of Florida Research Foundation, Inc. v. General Electric Co., 916 F.3d 1363, 1367, 129 USPQ2d 1409, 1411-12 (Fed. Cir. 2019) (relying on specification’s description of the claimed analysis and manipulation of data as being performed mentally "‘using pen and paper methodologies, such as flowsheets and patient charts’") The best paths as disclosed in the specification as performed the models to be trained is similar as a process to the mental process found in the court to be analogous to the mapping process or flow sheets. (MPEP 2106.04 (a) (2) section III B and C3). The claimed structures (primary/secondary models) are generic computer components and tools to perform the mental processes. The computer components are recited at a high level of generality and merely automates functions that could reasonable be performed using mental concepts, therefore acting as a generic computer to perform the abstract idea. Additional evidence includes:
EP 3675008 A1 by Ma et al; US Pub No. 2022/0294710 A1 by Swvigaradoss et al; US RE46153 E Makagon et al; US Patent No. 7,039,166 B1 by Peterson et al; US Patent No. 6,937,705 B1 by Godfrey et al; US Patent No. 6,898,277 B1 by Meteer et al; US Patent No. 6,823,054 B1 by Suhm et al;
The instant application, therefore, still appears to only implement the abstract ideas to the particular technological environments using what is generic components and functions in the related arts. The claim is not patent eligible.
The remaining dependent claims—which impose additional limitations—also fail to claim patent-eligible subject matter because the limitations cannot be considered statutory. In reference to claims 26-29 these dependent claim have also been reviewed with the same analysis as independent claim 21. Dependent claim 26 is directed toward weighting the distance function according to user satisfaction, likelihood of agreement and time efficiencies where analyzing and outputting the distance between expected outcomes and desired outcomes– applying technology for use in analyzing and outputting outcomes for a business practice. Dependent claim 27 is directed toward generating measure of prediction uncertainty for expected outcomes associated with proposals communicated at steps (h) and (o)- applying technology to analyze data for a business practice. Dependent claims 28 is directed toward business and user related data content used for the analysis of a business process- limiting data acted upon and data analysis for a business practice. Dependent claim 29 is directed toward determining desirable outcome for a business practice according to customer satisfaction, lowest cost or least time to implement- applying technology to limit data analysis for a business practice
The dependent claim(s) have been examined individually and in combination with the preceding claims, however they do not cure the deficiencies of claim 21. Where all claims are directed to the same abstract idea, “addressing each claim of the asserted patents [is] unnecessary.” Content Extraction & Transmission LLC v. Wells Fargo Bank, Nat 7 Ass ’n, 776 F.3d 1343, 1348 (Fed. Cir. 2014). If applicant believes the dependent claims 26-29 are directed towards patent eligible subject matter, they are invited to point out the specific limitations in the claim that are directed towards patent eligible subject matter.
In reference to Claims 23 and 30-33:
STEP 1. Per Step 1 of the two-step analysis, the claims are determined to include a non-transitory computer readable medium, as in independent Claim 23 and the dependent claims. Such mediums fall under the statutory category of "manufacture." Therefore, the claims are directed to a statutory eligibility category.
STEP 2A Prong 1. The claimed invention is directed to an abstract idea without significantly more. Medium claim 23 recites operational process of a) analyzing data comprising user profile b) determining desirable outcome of interaction between agent and equipment c) determining user state d) training a model e) initiating a negotiation between user and agent f) simulating plurality of proposals g) selecting best expected outcome h) communicating proposal associated with best outcome i) obtaining a response from user equipment j) updating data k) deriving a secondary RL model (l) determining whether response indicates user agreed to proposal m) refining first plausible action in response to suer has not agreed to proposal o) selecting a second plurality of proposals p) repeating steps h) -o) iteratively and based on distance between expected outcomes and desirable outcomes threshold terminate negotiation, fixed iterations reached conclude method or user agrees to proposals.
The claimed limitations which under its broadest reasonable interpretation, covers performance of analyzing data for a business process. When considered as a whole the claimed subject matter in light of the specification, the subject matter is directed toward analyzing data; determining outcome; training models; determining user state; performing simulations, selecting outcomes; receiving response; updating data, state, model; determining whether desirable outcome has been obtained; determining user behavior in real time, deriving of secondary RL model, performing simulation of a further step of the interactions by generating second expected outcomes, selecting a further action and communication of proposals/results is directed toward determining opportunities for deeper and more personalized interactions in a business environment. The specification discloses the “personalizing of the first model” a personalized interaction with a user to obtain a desirable outcome (para 0013) where the system delivers services, resolved customer issues provides remote learning with personalized treatment (para 0026). This allows the system to use a procedure to a personalized action path based on user historical and profile data (para 0028). The specification discloses the first/primary model is used as a reference/starting model from which a secondary model can be derived using additional optimization/domain strategies. The secondary model optimized/personalized for a singular user, the users state or expected experience results from decisions which makes clear that the focus of the invention as a whole is directed toward commercial interactions. Such concepts can be found in the abstract category of commercial interactions and marketing. These concepts are enumerated in Section I of the 2019 revised patent subject matter eligibility guidance published in the federal register (84 FR 50) on January 7, 2019) is directed toward abstract category of methods of organizing human activity.
STEP 2A Prong 2: The identified judicial exception is not integrated into a practical application because the claims fail to provide indications of patent eligible subject matter that integrate the alleged abstract idea into a practical application. The additional elements recited in the claim beyond the abstract idea include a non-transitory machine readable medium comprising executable instructions executed by a processing system to perform the operations according to the instructions.
The RL model does not perform any of the recited steps of the method.
The processing system applied to perform the operation “obtaining a response…”, “communicating the proposal…”, and “obtaining a response” which according to MPEP 2106.05(d) II (see also MPEP 2106.05(g)) the courts have recognized the following computer functions are claimed in a merely generic manner (e.g., at a high level of generality) where technology is merely applied to perform the abstract idea or as insignificant extra-solution activity.
Receiving or transmitting data over a network, e.g., using the Internet to gather data, Symantec, 838 F.3d at 1321, 120 USPQ2d at 1362 (utilizing an intermediary computer to forward information); TLI Communications LLC v. AV Auto. LLC, 823 F.3d 607, 610, 118 USPQ2d 1744, 1745 (Fed. Cir. 2016) (using a telephone for image transmission); OIP Techs., Inc., v. Amazon.com, Inc., 788 F.3d 1359, 1363, 115 USPQ2d 1090, 1093 (Fed. Cir. 2015) (sending messages over a network); buySAFE, Inc. v. Google, Inc., 765 F.3d 1350, 1355, 112 USPQ2d 1093, 1096 (Fed. Cir. 2014) (computer receives and sends information over a network); but see DDR Holdings, LLC v. Hotels.com, L.P., 773 F.3d 1245, 1258, 113 USPQ2d 1097, 1106 (Fed. Cir. 2014)
The additional element “processing system” applied to perform the operations “analyzing …data…”, “determining…desirable outcome…”, “determining…a user state…”, where the functions are performed by the processing system used for data analysis recited at a high level without any technical details applied to provide expected outcomes.
The additional element “processing system” applied to perform the operations “initiating ….negotiation”, where the steps performed by the processing system used for selecting business related actions recited at a high level without any technical details applied to provide expected outcomes.
The additional element “processing system” applied to perform the operations “training …a primary reinforcement learning (RL) model…” and “deriving of the secondary RL model” where the operation performed by the processing system used for training /deriving models, the limitations and specification do not provide any details as to how the training is implemented and therefore is so broad as to include any known means using generic programming. The specification discloses ¶ 0034 “the RL model can be trained to apply weighting factors (e.g. give greater weight to more recent interaction data). In another embodiment, models for a population, a population segment, and/or an individual user can be combined to construct a hybrid model”, ¶ 0039 “procedure 203 for training a reinforcement learning model and performing run-time simulations of proposed actions in an interaction with a user, in accordance with embodiments of the disclosure. The RL model is trained 231 using user features and proposals/actions to predict an outcome ( e.g. a response from the user). In an embodiment, the model is trained to map the most recent user state and a proposed action to an outcome for that action.”; ¶ 0045 …”The system then builds and trains a reinforcement learning (RL) model to map the user state and a system action (e.g. a proposal to be presented to the user) to an expected outcome (step 2408). This model may be trained using data from an individual user, a segment of similar users, or a population of users….” and wherein ¶ 0098 describes training data which can be acted upon. However the specification is silent with respect the technical details or process for “training” RL models focusing only on the function and outcome. The additional element “processing system” applied to perform the step “simulating …proposals” and “repeating steps (h)-(o)” where the steps performed by the processing system used for simulating steps in a business activity recited at a high level without any technical details applied to provide expected outcomes.
The limitations are broad and lack any technical details and therefore can be performed by any known means using generic programming to implement and perform the abstract idea. Furthermore the limitations are directed not toward technology but rather data collection and analysis for a business practice.
The specification discloses such simulations as generating expected outcomes for each of a plurality of possible actions which is mere automation of data analysis by repeatedly analyzing each possible action - .(MPEP 2106.05(d) II (ii) Bancorp Services v. Sun Life, 687 F.3d 1266, 1278, 103 USPQ2d 1425, 1433 (Fed. Cir. 2012) ("The computer required by some of Bancorp’s claims is employed only for its most basic function, the performance of repetitive calculations, and as such does not impose meaningful limits on the scope of those claims.
The functions are directed toward performing analysis on human behavior and commercial activity. The claim limitations use the model for data analysis and provides no details as to technical implementation. This is also true with respect to the “training” and/or “deriving” functions of the first and second models respectively. The limitations recite the training limitation so broad as to encompass simple programming or any and all means for training models to perform expected activities. The limitations do not delineate how the “training”, the “deriving” of the models achieves an improvement to technology or part of a solution to a problem rooted in existing technology. The only thing the claims discloses is about the use of the model in a commercial environment The functions are is recited at a high-level of generality such that it amounts to no more than applying the exception using generic computer components. Taking the claim elements separately, the operation performed by the Method at each step of the process is purely in terms of results desired and devoid of implementation of details. This is true with respect to the limitations “training a model” and “deriving a model” and the modelling of the data, as the claimed limitations do not provide any technology to perform the recited functions. Technology is not integral to the process as the claimed subject matter is so high level that any generic programming could be applied and the functions could be performed by any known means. Furthermore, the claimed functions do not provide an operation that could be considered as sufficient to provide a technological implementation or application of/or improvement to this concept (i.e. integrated into a practical application).
When the claims are taken as a whole, as an ordered combination, the combination of limitations (a)-(b) and (c) are directed toward analyzing data and determining outcomes of interaction and user state which is for the purpose of a business practice. The combination of limitations (a)-(c) and (d) are directed toward training of a model for business proposal outcomes using data from limitations (a)-(c) applying technology for performing business proposal outcomes. The combination of limitations (a)-(d) and ( e) –(h) are directed toward initiating a negotiation, where proposals are simulated applying the limitations of (a)-(d) and the outcomes of the simulation are communicated for use in a business practice. The combination of limitations (i) –(k) are directed toward obtaining a response from the proposal of limitations (a)-(h) and updating the data for analysis and then deriving a secondary model for use in analyzing data- directed toward analyzing business data for use on communicating business proposals of the analysis of limitations. The combination of limitations (a)-(k) and (l)-(o) is directed toward determining user response to proposals and based on user response refine the plausible actions for expected outcome of the simulation that are then selected according to and then repeating limitations (h)-(o) iteratively until the analysis computes a distance between expected outcomes and desirable outcomes or user terminates negotiation, or fixed number of iterations reached or business rules require process to conclude or user agreed to proposals– which is merely applying the technology to analyze business data and output the result until a condition is met/determined. The combinations of parts is not directed toward any technical process or technological technique or technological solution to a problem rooted in technology. Accordingly, when the claims are taken as a whole, as an ordered combination, the combination of steps are not integrate the judicial exception into a practical application as the claim process fails to impose meaningful limits upon the abstract idea. This is because the claimed subject matter as a whole is directed modelling human behavior and refining the simulations- a process directed toward analyzing human activity. Therefore, the claimed limitations fail to provide additional elements or combination or elements to apply or use the judicial exception in a manner that imposes a meaningful limit on the judicial exception. The functions recited in the claims recite the concept of analyzing and determining data for analysis in a model, selecting actions based on simulation outcomes and updating data, user state and model to refine possible action which is a process directed toward analysis of human behavior and commercial activity where technology is merely being applied to implement the abstract idea.
Similar to the claim limitations, the specification lacks technical disclosure of the “training” and “deriving” of the RL model, instead only states that the model is trained. The specification discloses that the model is trained to map user state and a proposal to an expected outcome, where the model is “trained” on user population, on a segment of user population on an individual or combination of user and interaction type which focuses on the data applied for analysis rather than technical disclosure (para 0033, par 0045). The specification discloses the training of the model as performing simulation of proposed actions in an interaction with a user using user features, proposals/actions to predict an outcome (para 0039) which is an application of the model and not technical details, again lacking technical disclosure. The specification further states the system simulating possible next steps in the interaction to predict outcomes of different proposals/outcomes in the action space using the trained model and then the system choosing the action/proposal that results in best expected outcome (para 0046). The specification discloses employing classifiers explicitly trained (generic training data) as well as explicitly trained (observing user behavior, operator preferences, historical data and extrinsic data) where the classifier is used to determine predetermined criteria which benefit maximum number of subscribers.
The integration of elements do not improve upon technology or improve upon computer functionality or capability in how computers carry out one of their basic functions. The integration of elements do not provide a process that allows computers to perform functions that previously could not be performed. The integration of elements do not provide a process which applies a relationship to apply a new way of using an application. The instant application, therefore, still appears only to implement the abstract idea to the particular technological environments apply what generic computer functionality in the related arts. The steps are still a combination made to automated workflow to model a business process and does not provide any of the determined indications of patent eligibility set forth in the 2019 USPTO 101 guidance. The additional steps only add to those abstract ideas using generic functions, and the claims do not show improved ways of, for example, a particular technical function for performing the abstract idea that imposes meaningful limits upon the abstract idea. Moreover, Examiner was not able to identify any specific technological processes that goes beyond merely confining the abstract idea in a particular technological environment, which, when considered in the ordered combination with the other steps, could have transformed the nature of the abstract idea previously identified. Accordingly, this additional element does not integrate the abstract idea into a practical application because it does not impose any meaningful limits on practicing the abstract idea. The claim is directed to an abstract idea.
STEP 2B; The claim(s) does/do not include additional elements that are sufficient to amount to significantly more than the judicial exception because as discussed above with respect to concepts of the abstract idea into a practical application. The additional elements recited in the claim beyond the abstract idea include a non-transitory machine readable medium comprising executable instructions that, when executed by the processing system, facilitate performance of operations -–is purely functional and generic. Nearly every computer system will include a “processor” capable of performing the basic computer functions of “analyzing”, “determining”, ‘training”, “deriving”, “performing simulation”, “selecting”, “communicating”, “receiving”, “updating”, and “refining” steps required by the method claims. The claimed primary reinforcement learning (RL) model does not perform any of the limitations.
Taking the claim elements separately, the function performed by the computer at each step of the process is purely conventional. Using a processing system processor to train and derive models, and using such models to determine results of a user action or plurality of actions, perform simulations and refining possible actions of behavior ----are some of the most basic functions of a computer.
When the claims are taken as a whole, as an ordered combination, the combination of steps does not add “significantly more” by virtue of considering the steps as a whole, as an ordered combination. All of these computer functions are generic, routine, conventional computer activities that are performed only for their conventional uses. See Elec. Power Grp. v. Alstom S.A., 830 F.3d 1350, 1353 (Fed. Cir. 2016). Also see In re Katz Interactive Call Processing Patent Litigation, 639 F.3d 1303, 1316 (Fed. Cir. 2011) Absent a possible narrower construction of the terms “analyzing”, “determining”, “training”, “deriving”, “performing simulation”, “selecting”, “receiving”, “updating”, “refining” and “selecting”.. are functions can be achieved by any general purpose computer without special programming. None of these activities are used in some unconventional manner nor do any produce some unexpected result. Applicants do not contend they invented any of these activities. In short, each step does no more than require a generic computer to perform generic computer functions.
As to the data operated upon, "even if a process of collecting and analyzing information is 'limited to particular content' or a particular 'source,' that limitation does not make the collection and analysis other than abstract." SAP America, Inc. v. Invest Pic LLC, 898 F.3d 1161, 1168 (Fed. Cir. 2018). Considered as an ordered combination, the computer components of Applicant’s claimed functions add nothing that is not already present when the steps are considered separately. The sequence of data reception-analysis modification-transmission is equally generic and conventional. See Ultramercial, Inc. v. Hulu, LLC, 772 F.3d 709, 715 (Fed. Cir. 2014) (sequence of receiving, selecting, offering for exchange, display, allowing access, and receiving payment recited as an abstraction), Inventor Holdings, LLC v. Bed Bath & Beyond, Inc., 876 F.3d 1372, 1378 (Fed. Cir. 2017) (sequence of data retrieval, analysis, modification, generation, display, and transmission), Two-Way Media Ltd. v. Comcast Cable Communications, LLC, 874 F.3d 1329, 1339 (Fed. Cir. 2017) (sequence of processing, routing, controlling, and monitoring). The ordering of the steps is merely being implemented by generic computer devices. The analysis concludes that the claims do not provide an inventive concept because the additional elements recited in the claims do not provide significantly more than the recited judicial exception.
According to 2106.05 well-understood and routine processes to perform the abstract idea is not sufficient to transform the claim into patent eligibility. As evidence the examiner provides:
With respect to application of model- see Alice; In re Ferguson, 558 F.3d 1359, 1364, 90 USPQ2d 1035, 1039-40 (Fed. Cir. 2009).
The specification discloses:
[00064] With reference again to FIG. 4, the example environment can comprise a
computer 402, the computer 402 comprising a processing unit 404, a system memory 406
and a system bus 408. The system bus 408 couples system components including, but
not limited to, the system memory 406 to the processing unit 404. The processing unit
404 can be any of various commercially available processors. Dual microprocessors and
other multiprocessor architectures can also be employed as the processing unit 404.
[00065] The system bus 408 can be any of several types of bus structure that can
further interconnect to a memory bus (with or without a memory controller), a peripheral
bus, and a local bus using any of a variety of commercially available bus architectures.
The system memory 406 comprises ROM 410 and RAM 412. A basic input/output
system (BIOS) can be stored in a non-volatile memory such as ROM, erasable
programmable read only memory (EPROM), EEPROM, which BIOS contains the basic
routines that help to transfer information between elements within the computer 402, such as during startup. The RAM 412 can also comprise a high-speed RAM such as
static RAM for caching data.
[000100] Further, the various embodiments can be implemented as a method, apparatus
or article of manufacture using standard programming and/or engineering techniques to
produce software, firmware, hardware or any combination thereof to control a computer
to implement the disclosed subject matter. The term "article of manufacture" as used
herein is intended to encompass a computer program accessible from any computer readable
device or computer-readable storage/communications media. For example,
computer readable storage media can include, but are not limited to, magnetic storage
devices (e.g., hard disk, floppy disk, magnetic strips), optical disks (e.g., compact disk
(CD), digital versatile disk (DVD)), smart cards, and flash memory devices (e.g., card,
stick, key drive). Of course, those skilled in the art will recognize many modifications
can be made to this configuration without departing from the scope or spirit of the
various embodiments.
[000104] As employed herein, the term "processor" can refer to substantially any
computing processing unit or device comprising, but not limited to comprising, singlecore
processors; single-processors with software multithread execution capability; multicore
processors; multi-core processors with software multithread execution capability;
multi-core processors with hardware multithread technology; parallel platforms; and
parallel platforms with distributed shared memory. Additionally, a processor can refer to
an integrated circuit, an application specific integrated circuit (ASIC), a digital signal
processor (DSP), a field programmable gate array (FPGA), a programmable logic
controller (PLC), a complex programmable logic device (CPLD), a discrete gate or
transistor logic, discrete hardware components or any combination thereof designed to perform the functions described herein. Processors can exploit nano-scale architectures
such as, but not limited to, molecular and quantum-dot based transistors, switches and
gates, in order to optimize space usage or enhance performance of user equipment. A
processor can also be implemented as a combination of computing processing units.
With respect to the limitations “training” model, the specification discloses
[0033]… In another embodiment, a primary RL model can be used as a starting or reference model from
which a secondary model can be derived through methods utilizing additional optimization or domain adaptation strategies. These secondary models may have the same functional capabilities as the primary RL model; alternatively, the secondary models may be further optimized or personalized for a singular user, the user's state, or
an expected experience resulting from decisions in the action space.
[0034]… “the RL model can be trained to apply weighting factors (e.g. give greater weight to more recent interaction data). In another embodiment, models for a population, a population segment, and/or an individual user can be combined to construct a hybrid model”,
[0039]… procedure 203 for training a reinforcement learning model and performing run-time simulations of proposed actions in an interaction with a user, in accordance with embodiments of the disclosure. The RL model is trained 231 using user features and proposals/actions to predict an outcome ( e.g. a response from the user). In an embodiment, the model is trained to map the most recent user state and a proposed action to an outcome for that action.”;
[0045]… “The system then builds and trains a reinforcement learning (RL) model to map the user state and a system action (e.g. a proposal to be presented to the user) to an expected outcome (step 2408). This model may be trained using data from an individual user, a segment of similar users, or a population of users….”
Paragraph 0098 describes training data which can be acted upon.
The specification discloses the “personalizing of the first model” a personalized interaction with a user to obtain a desirable outcome (para 0013) where the system delivers services, resolved customer issues provides remote learning with personalized treatment (para 0026). This allows the system to use a procedure to a personalized action path based on user historical and profile data (para 0028). The specification discloses the first/primary model is used as a reference/starting model from which a secondary model can be derived using additional optimization/domain strategies. The secondary model optimized/personalized for a singular user, the users state or expected experience results from decisions. The specification lacks technical disclosure of the “deriving” of the secondary model, instead only states that the secondary model can be derived using optimization/domain strategies. The specification discloses the “domain” to be business/rules preferences or domain expertise as constraints on the action space and/or input in selecting the next action. The domain experts can design a set of desirable action paths and the system can determine at each step of the interaction the best path for the user (para 0029). The specification discloses that the model is trained to map user state and a proposal to an expected outcome, where the model is “trained” on user population, on a segment of user population on an individual or combination of user and interaction type which focuses on the data applied for analysis rather than technical disclosure (para 0033, par 0045). The specification discloses the training of the model as performing simulation of proposed actions in an interaction with a user using user features, proposals/actions to predict an outcome (para 0039) again lacking technical disclosure. The specification further states the system simulating possible next steps in the interaction to predict outcomes of different proposals/outcomes in the action space using the trained model and then the system choosing the action/proposal that results in best expected outcome (para 0046). The specification discloses employing classifiers explicitly trained (generic training data) as well as explicitly trained (observing user behavior, operator preferences, historical data and extrinsic data) where the classifier is used to determine predetermined criteria which benefit maximum number of subscribers. The specification lacks technical details on the classifier functional process. The examiner therefore, disagrees that the use of the ML models as claimed are processes which cannot reasonably be performed using generic processes. According to Electric Power Group decision, the “selecting information, by content or source, for collection, analysis, and display does nothing significant to differentiate a process from ordinary mental processes, whose implicit exclusion from § 101 undergirds the information-based category of abstract ideas.” The courts distinguish between claims that recite mental processes performed by humans and claims that recite mental processes performed on a computer. As the Federal Circuit has explained "[c]ourts have examined claims that required the use of a computer and still found that the underlying, patent-ineligible invention could be performed via pen and paper or in a person’s mind." Versata Dev. Group v. SAP Am., Inc., 793 F.3d 1306, 1335, 115 USPQ2d 1681, 1702 (Fed. Cir. 2015). See also Intellectual Ventures I LLC v. Symantec Corp., 838 F.3d 1307, 1318, 120 USPQ2d 1353, 1360 (Fed. Cir. 2016). The use of a physical aid (e.g., pencil and paper or a slide rule) to help perform a mental step (e.g., a mathematical calculation) does not negate the mental nature of the limitation, but simply accounts for variations in memory capacity from one person to another. For instance, in CyberSource, the court determined that the step of "constructing a map of credit card numbers" was a limitation that was able to be performed "by writing down a list of credit card transactions made from a particular IP address." In making this determination, the court looked to the specification, which explained that the claimed map was nothing more than a listing of several (e.g., four) credit card transactions. The court concluded that this step was able to be performed mentally with a pen and paper, and therefore, it qualified as a mental process. 654 F.3d at 1372-73, 99 USPQ2d at 1695. See also Flook, 437 U.S. at 586, 198 USPQ at 196 (claimed "computations can be made by pencil and paper calculations"); University of Florida Research Foundation, Inc. v. General Electric Co., 916 F.3d 1363, 1367, 129 USPQ2d 1409, 1411-12 (Fed. Cir. 2019) (relying on specification’s description of the claimed analysis and manipulation of data as being performed mentally "‘using pen and paper methodologies, such as flowsheets and patient charts’") The best paths as disclosed in the specification as performed the models to be trained and derived is similar as a process to the mental process found in the court to be analogous to the mapping process or flow sheets. (MPEP 2106.04 (a) (2) section III B and C3). The claimed structures (primary/secondary models) are generic computer components and tools to perform the mental processes. The computer components are recited at a high level of generality and merely automates functions that could reasonable be performed using mental concepts, therefore acting as a generic computer to perform the abstract idea. Additional evidence includes:
EP 3675008 A1 by Ma et al; US Pub No. 2022/0294710 A1 by Swvigaradoss et al; US RE46153 E Makagon et al; US Patent No. 7,039,166 B1 by Peterson et al; US Patent No. 6,937,705 B1 by Godfrey et al; US Patent No. 6,898,277 B1 by Meteer et al; US Patent No. 6,823,054 B1 by Suhm et al;
The instant application, therefore, still appears to only implement the abstract ideas to the particular technological environments using what is generic components and functions in the related arts. The claim is not patent eligible.
The remaining dependent claims—which impose additional limitations—also fail to claim patent-eligible subject matter because the limitations cannot be considered statutory. In reference to claims 30-33 these dependent claim have also been reviewed with the same analysis as independent claim 23. Dependent claim 30 is directed toward weighting the distance function according to user satisfaction, likelihood of agreement and time efficiencies where analyzing and outputting the distance between expected outcomes and desired outcomes– applying technology for use in analyzing and outputting outcomes for a business practice. Dependent claim 31 is directed toward generating measure of prediction uncertainty for expected outcomes associated with proposals communicated at steps (h) and (o)- applying technology to analyze data for a business practice. Dependent claims 32 is directed toward business and user related data content used for the analysis of a business process- limiting data acted upon and data analysis for a business practice. Dependent claim 33 is directed toward determining desirable outcome for a business practice according to customer satisfaction, lowest cost or least time to implement- applying technology to limit data analysis for a business practice
The dependent claim(s) have been examined individually and in combination with the preceding claims, however they do not cure the deficiencies of claim 23. Where all claims are directed to the same abstract idea, “addressing each claim of the asserted patents [is] unnecessary.” Content Extraction & Transmission LLC v. Wells Fargo Bank, Nat 7 Ass ’n, 776 F.3d 1343, 1348 (Fed. Cir. 2014). If applicant believes the dependent claims 30-33 are directed towards patent eligible subject matter, they are invited to point out the specific limitations in the claim that are directed towards patent eligible subject matter.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure. US Pub No. 2022/0197306 A1 by Cella et al; US Pub No. 2010/0131916 A1 by Prigge
Any inquiry concerning this communication or earlier communications from the examiner should be directed to MARY M GREGG whose telephone number is (571)270-5050. The examiner can normally be reached M-F 9am-5pm.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Christine Behncke can be reached at 571-272-8103. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of published or unpublished applications may be obtained from Patent Center. Unpublished application information in Patent Center is available to registered users. To file and manage patent submissions in Patent Center, visit: https://patentcenter.uspto.gov. Visit https://www.uspto.gov/patents/apply/patent-center for more information about Patent Center and https://www.uspto.gov/patents/docx for information about filing in DOCX format. For additional questions, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/MARY M GREGG/Examiner, Art Unit 3695
Prosecution Insights