DETAILED ACTION
This non-final office action is responsive to application 17/794,307 as filed 07/21/2022.
Claim status is Preliminary Amendment submitted 07/26/22 where claims 1-21 are pending and under examination; claim 1 being of independent form; claim 8-9 are currently amended; claims 10-21 are newly presented.
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 .
Priority
Receipt is acknowledged of certified copies of papers required by 37 CFR 1.55.
Information Disclosure Statement
As required by MPEP 609, the applicant’s submissions of the Information Disclosure Statement dated 07/22/22 is acknowledged by the examiner and the cited references have been considered in the examination of the claims now pending. As required by MPEP 609, a copy of the PTOL-1449 initialed and dated by the examiner is attached to the instant office action.
Drawings
The drawings are not of sufficient quality to permit examination. Accordingly, replacement drawing sheets in compliance with 37 CFR 1.121(d) are required in reply to this Office action. The replacement sheet(s) should be labeled “Replacement Sheet” in the page header (as per 37 CFR 1.84(c)) so as not to obstruct any portion of the drawing figures. If the changes are not accepted by the examiner, the applicant will be notified and informed of any required corrective action in the next Office action.
Applicant is given a shortened statutory period of TWO (2) MONTHS to submit new drawings in compliance with 37 CFR 1.81. Extensions of time may be obtained under the provisions of 37 CFR 1.136(a) but in no case can any extension carry the date for reply to this letter beyond the maximum period of SIX MONTHS set by statute (35 U.S.C. 133). Failure to timely submit replacement drawing sheets will result in ABANDONMENT of the application.
Figure 2 should be designated by a legend such as --Prior Art-- because only that which is old is illustrated, see MPEP § 608.02(g). Corrected drawings in compliance with 37 CFR 1.121(d) are required in reply to the Office action to avoid abandonment of the application. The replacement sheet(s) should be labeled “Replacement Sheet” in the page header (as per 37 CFR 1.84(c)) so as not to obstruct any portion of the drawing figures. If the changes are not accepted by the examiner, the applicant will be notified and informed of any required corrective action in the next Office action. The objection to the drawings will not be held in abeyance.
Figure 2 additionally must comply with CFR 1.84 for the reason that Figure 2 depicts “© 1999 Encyclopaedia Britannica, Inc.” shown bottom left. Inclusion of copyright notice is permitted only if authorization language set forth in CFR1.71(e) is included at the beginning of the specification.
Specification
The lengthy specification, at 191 pages, has not been checked to the extent necessary to determine the presence of all possible minor errors. Applicant’s cooperation is requested in correcting any errors of which applicant may become aware in the specification.
Claim Objections
Claims 1-21 are objected to because of the following informalities:
Claim 1 is replete with contingent limitations, notedly terms “in case of”, “if”, and “if any” are used throughout the claims. These terms may not be further limiting because the case may (not) be, or not-if/while, all-else, else-if. Remaining claims depend from the contingent limitations and add further contingencies without resolving the issue. See MPEP 2111.04.
Claim 2 limitation of “the compliant constraint” should conclude with the use of semi-colon as a break to separate from the following indented limitation.
Claim 4 fails to end with a period so as to be in proper form of claim, see MPEP 608.01(m).
Appropriate correction is required.
Claim Rejections - 35 USC § 112
The following is a quotation of 35 U.S.C. 112(b):
(b) CONCLUSION.—The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the inventor or a joint inventor regards as the invention.
Claims 2-3, 10 and 16 are rejected under 35 U.S.C. 112(b), as being indefinite for failing to particularly point out and distinctly claim the subject matter which the inventor or a joint inventor, regards as the invention. Particularly, claim 2 recites “behaviors that should be prohibited” in final limitation. The phrase “should be” renders the claim indefinite because it is unclear whether the limitation following the phrase are part of the claimed invention. Claims 10 and 16 depend from claim 2 without correcting the issue. Additionally, claim 3 recites “the isolation condition” for which there is insufficient antecedent basis. Therefore, claims 2-3, 10 and 16 are rejected as indefinite under §112(b). See MPEP 2173.05(d).
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.
13. Claims 8-21 are rejected under 35 U.S.C. 101 for being directed to non-statutory subject matter. Claim 8 and 9 are drawn to a respective system and device that adopt the method of claim 1, similarly claims 10-21 respectively adopt method claims 2-7 which are dependent from method claim 1. The system and device use modules which under the broadest reasonable interpretation can be a software system implemented by computer programs which amounts to software per-se. It fails to assert that the software system and device are executed by processor and recorded on a non-transitory computer-readable medium so as to be structurally and functionally interrelated to the medium and permit the function of the descriptive material to be realized. A computer program is merely a set of instructions capable of being executed by a computer. Without a processor and computer-readable medium to realize the computer program's functionality, the computer program constitutes non-statutory functional descriptive material. See MPEP 2106.03.
Claims 1-21 are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more. In determining whether the claims are subject matter eligible, the examiner applies guidance set forth under MPEP 2106.
Step 1: Is the claim to a process, machine, manufacture, or composition of matter? Yes—all claims fall within (or could be amended to fall within) one of the four statutory categories: claims 1-7 are a method/process, claims 8 and 10-15 are a system/machine, claims 9 and 16-21 are a device/article of manufacture. In the case of claims 8-21, if a claim could be amended to fall within one of the statutory categories, then the analysis should proceed. Thus, the analysis should continue per MPEP 2106.03.
Step 2A, prong one: Does the claim recite an abstract idea, law of nature or natural phenomenon? Yes—the claims, under the broadest reasonable interpretation, recites an abstract idea. In this case, claims fall within the enumerated grouping of abstract idea being “Mental Processes.” Particularly, claims recite:
“establishing a safety range of operation data” (Mental judgment, e.g. set speed limits)
“setting a constraint condition and a heuristic end condition, and setting an emergency plan for the constraint condition” (Mental judgment, e.g. if-fire, then-evacuate till-safe)
“performing a heuristic process: acquiring current basic working condition data, operation data and an emergency plan, wherein the operation data comprises at least one operation data dimension, and in case of no emergency plan, only the operation data is acquired; selecting at least one operation data dimension from the operation data by means of a random algorithm, randomly generating a value within a safety range of the selected operation data dimension to form new operation data of the selected operation data dimension,
“checking the constraint condition; if the constraint condition is not met, starting the emergency plan, if any” (Mental judgment e.g. if-yes/no, then do-routine, mitigate risk)
“performing, after a working condition is stable, self-learning on the basic working data, the new operation data and evaluation data generated therefrom if the heuristic working state is not changed” (Mental evaluation, e.g. self-educating performed cognitively on new info)
“if the heuristic end condition is not triggered, performing a next heuristic process; or if the heuristic end condition is triggered, ending the heuristic self-learning state” (Mental judgment, e.g. if-yes/no, then-proceed/terminate)
Focus of the claim concerns a heuristic process with an emergency plan and a safety range. When read in light of the “Detailed Description” [0061] only a single paragraph is given that isn’t drawn to 1 of 97 embodiments, none of which appear to strongly correlate with claim 1, and the preceding brief summary [0006-41] largely repeats claim language. Summary introduces per [0004,16] “humans” and [0022] “emergency plan comprises… alarm is triggered.” A definition for heuristic is offered by OpenTrain AI [P.1 ¶1] “heuristic is a strategy or method that guides the search for solutions… by making educated guesses or applying practical rules of thumb” and/or Colombe [P.1 ¶1] “The heuristic approach, which dominated for the first several decades of AI research, depends entirely upon recipes for solving data-processing and decision problems that have been thought out and encoded by human programmers.” In view of this, such humans are capable of performing heuristic process for emergency plan with safety range. Ordinarily, when a human contacts emergency services like police, the first question is: What is the nature of your emergency? Similarly, the nature of an emergency plan and safety range are not precluded from mental performance merely because conditions and constraints are broadly applied, or can be triggered by alarm limit. Therefore, humans may practically perform such process as part of mental processes being the abstract idea.
Step 2A, prong two: Does the claim recite additional elements that integrate the judicial exception into a practical application? No—a practical application is not integrated by the judicial exception because the additional elements are as follows:
“automatically executing the new operation data by a device” MPEP 2106.05(f) merely uses a computer as a tool to perform an abstract idea
Preamble conveys machine learning, MPEP 2106.05(g) adding insignificant extra-solution activity to the judicial exception
Balance of the claim concerns device execution and machine learning. The device is recited at a high level of generality and the machine learning has no model e.g. neural network. Even if one were to interpret the machine learning more narrowly to include self-learning, this would not elevate the claim because numerous techniques like self-supervision, self-organizing maps, radial basis functions or reinforcement feedback are established in respective field of endeavor, thus failing to meaningfully limit the claim. No concrete or real-world application is set forth in the claim. Accordingly, the claim remains drawn to the judicial exception and the additional elements do not integrate the abstract idea into a practical application.
Step 2B: Does the claim recite additional elements that amount to significantly more than the judicial exception? No—the claims do not include additional elements that amount to significantly more than the judicial exception. As discussed above with respect to integration of the abstract idea in to a practical application, the additional elements are identified with respect to MPEP 2106.05 and do not demonstrate inventive concept. Particularly, additional elements are as follows:
“automatically executing the new operation data by a device” MPEP 2106.05(f) merely uses a computer as a tool to perform an abstract idea. More particularly, conventional computing components or general purpose devices do not satisfy the test of particular machine under MPEP 2106.05(b).
Preamble conveys machine learning, MPEP 2106.05(g) adding insignificant extra-solution activity to the judicial exception. More particularly, said extra-solution activity is a well-understood, routine and conventional (WURC) activity as evidenced by Chen et Zhang “Learning How to Self-Learn: Enhancing Self-Training using Neural Reinforcement Learning” arXiv: 1804.05734v1 at [P.1 Sect.1 ¶3] “Traditional self-training solutions are designed manually based on task-specific heuristics” similar [P.3 Sect.3 ¶3] “Traditional self-training approach usually requires human efforts to design specific confidence metrics based on the heuristics” emphasis being traditional i.e. conventional.
In view of the foregoing, these additional elements do not amount to significantly more. If the claim language provides only a result-oriented solution, with insufficient detail for how a computer accomplishes it, then the claims do contain an inventive concept. Taken alone, their additional elements do not amount to significantly more than the above-identified judicial exception (the abstract idea). Looking at the limitations as an ordered combination adds nothing that is not already present when looking at the elements taken individually. There is no indication that the combination of elements satisfies the tests of particular machine or particular transformation in a manner that meaningfully limits the claim through a technical solution. Their collective functions merely provide conventional computing device implementation.
For at least the above reasons, claim 1 is not patent eligible. Dependent claims when analyzed as a whole are held to be patent ineligible under 35 U.S.C. 101 because the additional limitations fail to establish that the claims are not directed to an abstract idea, or that they include additional elements which integrate the judicial exception into a practical application or amount to significantly more.
Dependent claim 2 discloses wherein the constraint condition comprises a precondition of optimization objective, a compliant constraint that violates national standards, and a negative list of operation data. The limitations are considered part of the abstract idea being mental processes. The precondition’ condition being fulfilled as an optimization objective is a mental evaluation, nothing in the claim points to what the pre/condition is or what the objective actually is or how it is optimized. The compliance appearing in “various” data that violates national standards is governed by bodies that set standards such as NIST, and the phrases hinder quality and negative influence are without substance of a technical nature. Further, the dangerous behavior prohibition considering personnel security is human centric hazard concern, and consideration of device is simply a generic computing component. The only additional element is “device” such as widget recited a high level of generality and amounts to mere use of general computer components to perform ab abstract idea under MPEP 2106.05(f) and which fails to satisfy the test of particular machine under MPEP 2106.05(b). The claim as a whole embellishes the abstract idea of mental processes and additional element does not integrate the judicial exception into a practical application or amount to significantly more.
Dependent claim 3 discloses isolation condition is stricter than constraint condition, and when the isolation condition is triggered, return to previous operation. The limitation is considered part of the abstract idea being mental process as evaluation and judgment. No formal definition is given for strict nor even the conditions themselves. A condition being triggered is simply met, return to prior operation can be judged as a decision point in plan or flowchart. There are no additional elements.
Dependent claim 4 discloses wherein the emergency plan has preset value and an alarm, the alarm is triggered when the emergency plan is started. This is considered part of the abstract idea being mental processes as judgment or opinion. For example, flag information deemed to render unsafe and which may be contingent on some index of a template. There are no additional elements.
Dependent claims 5 discloses wherein the heuristic condition is a coverage rate of basic data reaching a preset proportion. The term coverage rate has but singular mention in the specification which repeats claim language. The limitation is considered to be part of the abstract idea being mental processes as an evaluation or judgment. This could be implemented a common metric or measurement meeting some criteria for sufficiency of the data being considered. There are no additional elements.
Dependent claim 6 discloses wherein evaluation data comprises an optimization objective value or a restrictive result value. This embellishes evaluation which is considered part of the mental process as mental evaluation. The objective is not set forth nor how it is optimized or similarly how some result is considered restrictive, it merely conveys that data comprises values. There are no additional elements.
Dependent claim 7 recites wherein evaluation data is superior to data corresponds to other data under the same data, and data records are updated. This is considered part of the abstract idea being mental process of evaluating information and updated by hand with aid of template or pen and paper. For example, use best data to file changes to documents. There are no additional elements.
Dependent claim 8 discloses adopting the method according to claim 1 in a system comprising modules which transmit data between modules and pre-store data in addition to limitations already addressed. The modules in a system amount to mere instructions to implement an abstract idea on a computer or merely uses a computer as a tool to perform an abstract idea under MPEP 2106.05(f) and which does not qualify as a particular machine under MPEP 2106.05(b). Further, the function to perform transmitting and storing amounts to adding insignificant extra-solution activity under MPEP 2106.05(g) and which more particularly is a well-understood, routine and conventional (WURC) activity as set forth per MPEP 2106.05(d)(II)(i,iv) “The courts have recognized the following computer functions as well-understood, routine, and conventional functions when they are claimed in a merely generic manner (e.g., at a high level of generality) or as an insignificant extra-solution activity: i. Receiving or transmitting data over a network… iv. Storing and retrieving information in memory.” Therefore, the additional elements do not integrate the judicial exception into a practical application or amount to significantly more.
Dependent claim 9 discloses adopting the method according to claim 1 in a device comprising modules which transmit data between modules and pre-store data in addition to limitations already addressed. The modules in a device amount to mere instructions to implement an abstract idea on a computer or merely uses a computer as a tool to perform an abstract idea under MPEP 2106.05(f) and which does not qualify as a particular machine under MPEP 2106.05(b). Further, the function to perform transmitting and storing amounts to adding insignificant extra-solution activity under MPEP 2106.05(g) and which more particularly is a well-understood, routine and conventional (WURC) activity as set forth per MPEP 2106.05(d)(II)(i,iv) “The courts have recognized the following computer functions as well-understood, routine, and conventional functions when they are claimed in a merely generic manner (e.g., at a high level of generality) or as an insignificant extra-solution activity: i. Receiving or transmitting data over a network… iv. Storing and retrieving information in memory.” Therefore, the additional elements do not integrate the judicial exception into a practical application or amount to significantly more.
Dependent claims 10-21 disclose adopting the method of claims 2-7 in the system and device with the body of claims 8-9, respectively. The limitations of method claims 2-7 are already addressed above as serving to embellish the abstract idea, and the system/device of claim 8-9 already addressed above as implementing the abstract idea on general computer with WURC functions. The combination does not demonstrate inventive concept nor do the additional elements do not integrate the judicial exception into a practical application or amount to significantly more.
Taken alone, their additional elements do not amount to significantly more than the above-identified judicial exception (the abstract idea). Looking at the limitations as an ordered combination adds nothing that is not already present when looking at the elements taken individually. There is no indication that the combination of elements satisfies the tests of particular machine or particular transformation in a manner that meaningfully limits the claim through a technical solution. Their collective functions merely provide conventional computer implementation.
Claim Rejections - 35 USC § 103
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
The factual inquiries for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
1. Determining the scope and contents of the prior art.
2. Ascertaining the differences between the prior art and the claims at issue.
3. Resolving the level of ordinary skill in the pertinent art.
4. Considering objective evidence present in the application indicating obviousness or nonobviousness.
Claims 1 and 8-9 are rejected under 35 U.S.C. 103 as being unpatentable over:
Gooch, Arthur, US PG Pub No 2020/0192340A1 hereinafter Gooch, in view of
Cha et al., US PG Pub No 2019/0258566A1 hereinafter Cha, and further in view of
Bar, Schirin, PCT WO2021/0525289A1 hereinafter Bar (corresponds US2022/0374002A1).
With respect to claim 1, Gooch teaches:
A machine heuristic learning method for operation behavior record management {Gooch Fig 1 introduced summary [0005] “method, performed by one or more data processing apparatus, for training an industrial plant controller… reinforcement learning technique” comprises [0071] “database management” and a common heuristic is ϵ-greedy exploration strategy [0048]}, comprising:
establishing a safety range of operation data; setting a constraint condition and a heuristic end condition {Gooch discloses [0051] “safe operating regimes (e.g., range of safe temperature values …safety factor”, end condition is [0058] “a set of ‘goal’ state vectors characterizing acceptable states of the industrial plant” and a constraint condition is [0058] “threshold distance of (or identical to) one or more of the goal state vectors” i.e. acceptable state subject to simulated test events of industrial plant, modeled by reinforcement learning (RL) with a reward Fig 1-2. The reward drives RL as a basis for heuristic which more particularly is [0048] “epsilon-greedy exploration strategy”}, and
setting an emergency plan for the constraint condition {Gooch Fig 4:404 “policy” is plan and emergency is [0046] “equipment failure (e.g., a valve breaking)… simulate the failure of a valve” such that the [0052] “goal state vector may correspond to the industrial plant being shut down” similarly at [0058] testing/simulated events, industrial plant is constrained by equipment failure in real terms, see also [0061-63]};
performing a heuristic process: acquiring current basic working condition data, operation data and an emergency plan, and in case of no emergency plan, only the operation data is acquired {Gooch [0039] “logged data characterizing the actual operation of the industrial plant” [0005] “input comprising a state vector characterizing a state of the industrial plant at the current time step” Fig 1:100 ind plant, state entails condition e.g. valve is broken or open/closed positions [0046], Fig 4:404 policy is plan for failure of equipment, current is at time-step and operation data is logged}; automatically executing the new operation data by a device, and entering a heuristic working state {Gooch [0017-18] execution by device is disclosed, more particularly [0040] iterative adjustment based on training data for controller that controls selected actions or operations for industrial plant shown Fig 1-top loop of state-action and training system described [0037] “continuous operation of the industrial plant… generate training data” with state shown Figs 1-4 and heuristic being the exploration strategy [0048], [0022]};
checking the constraint condition; if the constraint condition is not met, starting the emergency plan if any {Gooch [0059-57] determining within threshold and certification tests convey checking constraints conditions, shutting an industrial plant down addresses emergency and doing so as an action selected by controller policy is a plan Fig 4:404, [0056,62-63]};
if the heuristic end condition is not triggered, performing a next heuristic process; or, if the heuristic end condition is triggered, ending the heuristic self-learning state {Gooch Figs 2:212, 3:306 “termination criterion satisfied?” is ending, described [0052-53] “determine the termination criterion has been met for the simulated trajectory when the subsequent state vector characterizing the simulated state of the industrial plant after the control action is performed is within a threshold distance of (or is identical to) a predetermined goal state” end/goal state relative to simulated state determined, see also [0057,59] terminates training for final output thus modeling process with heuristic is ended; repeating steps is also disclosed to convey performing next process. Claim uses ‘or’ in the alternative}.
However, Gooch does not expressly disclose dimension nor selection thereof which is disclosed by Cha:
wherein the operation data comprises at least one operation data dimension; selecting at least one operation data dimension from the operation data by means of a random algorithm, randomly generating a value within a safety range of the selected operation data dimension to form new operation data of the selected operation data dimension {Cha [0059,58] “each dimension of each parameter vector can be randomly selected from a sample space (which initially may be [-1,1])” so as to [0062,69] “reduce the sample space for each dimension” using [0074-76] “random sampling algorithm” similar [0046,66]. The range of safety is thus [0046] “set to be real numbers between -1 and 1 ([-1,1])” e.g. Fig 4 shows 0.3 between [-1,1] and Fig 9:S220 “Randomly generate first parameter vectors having same dimension” iterated in loop with features representing data for testing, e.g. Figs 8 and 3 show algorithmic implementation},
Cha is directed to heuristic processes for data management with vector-based algorithms thus being analogous. A person having ordinary skill in the art would have considered it obvious prior to the effective filing date to specify dimension and selection thereof per Cha for Gooch’s safety range in combination for the motivation to “reduce the sample space for each dimension” (Cha [0069]) and/or “automatically generate a search heuristic optimized to any arbitrary program and can offer a consistent test performance (outputting maximum branch coverage) for any arbitrary program” (Cha [0040]).
However, Gooch in combination does not explicitly describe the RL with reward characterized as “self” learning which is disclosed by Bar:
performing, after a working condition is stable, self-learning on the basic working data, the new operation data and evaluation data generated therefrom if the heuristic working state is not changed {Bar [P.5 Lines4-19] “self-learning system for online scheduling, where RL agents are trained” illustrated Fig 1, [P.7 Lines2-32] “the RL agent observes the new state and reward as an evaluation” and operation of manufacturing plant shown Fig 3 or described [P.5 Line24], [P.1 Lines3-18] CNC machines. Further, [P.2 Lines6-15] discloses heuristics, rules and constraints – constraints or criteria shown Fig 3 optimization criteria can be user defined. A working condition being stable may include “discrete event” [P.4 Line26], additionally the learning is noted “until…converge” [P.8 Lines37-38] such that learning is considered stable if it is converged and may provide ending criteria, additionally the training may occur online or offline [P.12 Lines24-25]}; and
Bar is directed to methods of operation information management for manufacturing machines with machine learning thus being analogous. A person having ordinary skill in the art would have considered it obvious prior to the effective filing date to perform self-learning per Bar’s RL for RL of Gooch in combination to arrive at the invention as claimed to establish that RL is indeed self-learning, thus supporting Gooch’s teaching of RL as a type of self-learning, and/or for the motivation being that “A self-learning product routing system would reduce the engineering effort, as the system learns the decision for many situations by itself” (Bar [P.2 Line38 – P.3 Line2]).
With respect to claim 8, the rejection of claim 1 is incorporated. The difference in scope being a system adopting the method of claim 1, comprising modules which transmit data between the modules and pre-store data in addition to the limitations of claim 1. Gooch discloses [0004] “system implemented as computer programs” where [0070-72] “computer programs, i.e., one or more modules of computer program instructions encoded… encode information for transmission” and “store one or more modules …distributed across multiple sites and interconnected by a data communication network” Fig 5 shows computer environment as would be appreciated by the skilled artisan. Accordingly, the rejection of claim 1 is applied to claim 8.
With respect to claim 9, the rejection of claim 1 is incorporated. The difference in scope being a device adopting the method of claim 1, comprising devices which transmit data between devices and which pre-store data in addition to limitations of claim 1. Gooch discloses [0071] “’data processing apparatus’ refers to data processing hardware and encompasses all kinds of apparatus, devices, and machines for processing data” and [0068-69] “software, firmware, hardware, or a combination… cause the apparatus to perform the operations” illustrated Fig 5 computer environment as would be appreciated by the skilled artisan. Accordingly, the rejection of claim 1 is applied to claim 8.
Claims 2, 10 and 16 are rejected under 35 U.S.C. 103 as being unpatentable over Gooch, Cha and Bar in view of:
Mohseni et al., “Practical Solutions for Machine Learning Safety in Autonomous Vehicles” hereinafter Mohseni (arXiv: 1912.09630v1).
With respect to claim 2, the combination of Gooch, Cha and Bar teaches the machine heuristic learning method for operation behavior record management according to claim 1. Mohseni teaches wherein the constraint condition comprises a precondition of an optimization objective, a compliant constraint, and a negative list of operation data;
the precondition of the optimization objective means that the optimization objective is fulfilled under the condition of meeting the precondition {Mohseni [P.2 ¶5] “minimize the risk to an acceptable level” acceptable is meeting the precondition, risk is objective optimized by minimization similar at [P.2 ¶1,6-7]. See also [Sect.3 ¶1-3] safe fail and safety margin};
the compliant constraint refers to a case, appearing in various result evaluation data and caused by the basic working condition data and operations, that violates national standards, hinders the quality of products from reaching the standard, and has a negative influence on a subsequent process {Mohseni [P.2 ¶2-7] ISO 26262 & ISO/PAS 21448 international standards, violating is “below the acceptable level” i.e. substandard quality [P.5 ¶2] “evaluate our failure predictor model based on error” error-based evaluation may include false positive and false negative [P.3 Sect.2 Last2¶]. See Fig 3 [P.5 ¶5] “training reject options into any DNN… self-supervised learning”};
the negative list of the operation data refers to dangerous operation behaviors that should be prohibited out of consideration of device and personnel security {Mohseni [P.2 ¶3,7] “Hazard Analysis and Risk Assessment (HARA) to determine vehicle level hazards… applied to autonomous driving” and “identifying the hazards due to inadequate performance functionality, insufficient situational awareness, reasonably foreseeable misuse or shortcomings of human machine interface” so as to “mitigate hazards… Safeguards for non-experts end-users of autonomous vehicles” [P.4 ¶2-3]}.
Mohseni is directed to self-supervised learning for safety thus being analogous. A person having ordinary skill in the art would have considered it obvious prior to the effective filing date to use the teachings of Mohseni in combination to arrive at the invention as claimed for the motivation that “standards mandate a meticulous analysis of hazards and risks… hazards and risks guide the safety engineers towards safety goals that are then used to create functional safety requirements” [P.2 ¶2-3] further benefit including “first practical ML safety solution… trigger appropriate warnings” [P.4 ¶4] and “second practical solution for ML safety leverages robustness techniques to improve Safety Margins of ML models in autonomous vehicles” [P.5 ¶5].
With respect to claim 10, which recites adopting the method according to claim 2 as well as the full body of claim 8 system, the system is taught by the combination of Gooch, Cha and Bar, and further claim 2 is taught in combination with Mohseni. Therefore, the rejection of claims 8 and 2 with equal motivation are applied to claim 10.
With respect to claim 16, which recites adopting the method according to claim 2 as well as the full body of claim 9 device, the device is taught by the combination of Gooch, Cha and Bar, and further claim 2 is taught in combination with Mohseni. Therefore, the rejection of claims 9 and 2 with equal motivation are applied to claim 16.
Claims 3, 11 and 17 are rejected under 35 U.S.C. 103 as being unpatentable over Gooch, Cha and Bar in view of:
Cao et Tang, “A Reinforcement Learning Hyper-Heuristic in Multi-Objective Single Point Search with Application to Structural Fault Identification” hereinafter Cao (arXiv: 1812.07958v1).
With respect to claim 3, the combination of Gooch, Cha and Bar teaches the machine heuristic learning method for operation behavior record management according to claim 1. Cao teaches wherein
the isolation condition is stricter than the constraint condition {Cao [P.9] Eq.17 αl ≤ αi ≤ αu comparison operator ≤ is strictness with lower and upper bounds for the fault index αi thus fault/failure is isolated/identified within constraint of range, [P.8-9 Sect.5]}; and
when the isolation condition is triggered in the heuristic working state, it is necessary to return to a previous operation {Cao Fig 2 shows hyper-heuristic in loop of flowchart, the loop thus returns to a previous operations comprising stop criteria, re-seed and mosa/r algorithms detailed per [P.4 Sect.3] if-else conditions, and criteria [P.9 ¶2]}.
Cao is directed to machine learning with heuristics for fault identification thus being analogous. A person having ordinary skill in the art would have considered it obvious prior to the effective filing date to employ the teachings of Cao in combination to arrive at the invention as claimed as applying known techniques to known methods ready for improvement to yield predictable results and/or for the motivation “customizing the heuristic repository to meet different optimization requirements… solve a representative structural fault identification problem without using any domain knowledge, as the hyper-heuristic framework autonomously adjusts the search iteratively during search. Due to the adaptive nature of the proposed methodology, the newly proposed framework can be extended to a variety of design and manufacturing optimization applications” (Cao [P.11 ¶2]).
With respect to claim 11, which recites adopting the method according to claim 3 as well as the full body of claim 8 system, the system is taught by the combination of Gooch, Cha and Bar, and further claim 3 is taught in combination with Cao. Therefore, the rejection of claims 8 and 3 with equal motivation are applied to claim 11.
With respect to claim 17, which recites adopting the method according to claim 3 as well as the full body of claim 9 device, the device is taught by the combination of Gooch, Cha and Bar, and further claim 3 is taught in combination with Cao. Therefore, the rejection of claims 9 and 3 with equal motivation are applied to claim 17.
Claims 4, 12 and 18 are rejected under 35 U.S.C. 103 as being unpatentable over Gooch, Cha and Bar in view of:
Corso, Ronald, US Patent No 5,486,998 hereinafter Corso.
With respect to claim 4, the combination of Gooch, Cha and Bar teaches the machine heuristic learning method for operation behavior record management according to claim 1. Corso teaches wherein
the emergency plan comprises a preset value of the operation data and an alarm mode {Corso Figs 5A-D “Emergency” flowchart/plan with decision points and replete “alarm ranges” e.g. [Col21 Lines48-54] “alarm values: low-low≤low≤high≤high-high …alarm range results in both generating alarms and causing the supervisor to enter an emergency state” and operation data may regard a coal facility Fig 3A and/or training neural net Fig 3B:278}; and
when the emergency plan is started, the operation data is modified into the preset value, and an alarm is triggered {Corso Fig 7:732 “alarm activation rule” for alarms of emergency plan Figs 5A-D, [Col19 Lines45-46] “an emergency state 408 will be automatically entered if an alarm activates” and again [Col21 Lines48-54]. See also Figs 10:1264 retrain neural net, Fig 11B heuristic and [Col9 Line21] “range of 3-7% for safety”}
Corso is directed to emergency planning with safety range, machine learning and heuristics thus being analogous. A person having ordinary skill in the art would have considered it obvious prior to the effective filing date to specify emergency with alarms per Corso in combination to arrive at the invention as claimed for the motivation “advantage of this invention is to fully automate process alarm handling whereby an operator response to any alarm is not required for safely controlling the process” [Col8 Lines16-18] and/or “automatically execute alternative control process strategies, including emergency shut down” [Col2 Lines55-57] for example “shut down in 10 seconds to 90 seconds, depending on the urgency of the emergency” [Col17 Line59].
With respect to claim 12, which recites adopting the method according to claim 4 as well as the full body of claim 8 system, the system is taught by the combination of Gooch, Cha and Bar, and further claim 4 is taught in combination with Corso. Therefore, the rejection of claims 8 and 4 with equal motivation are applied to claim 12.
With respect to claim 18, which recites adopting the method according to claim 4 as well as the full body of claim 9 device, the device is taught by the combination of Gooch, Cha and Bar, and further claim 4 is taught in combination with Corso. Therefore, the rejection of claims 9 and 4 with equal motivation are applied to claim 18.
Claims 5, 13 and 19 are rejected under 35 U.S.C. 103 as being unpatentable over Gooch, Cha and Bar in view of:
Rapp et al., “On the Trade-off Between Consistency and Coverage in Multi-label Rule Learning Heuristics” hereinafter Rapp (arXiv: 1908.03032v1).
With respect to claim 5, the combination of Gooch, Cha and Bar teaches the machine heuristic learning method for operation behavior record management according to claim 1. Rapp teaches wherein
the heuristic end condition is that a coverage rate of the basic working condition data reaches a preset proportion {Rapp [P.4 Sect2.4] Eq.4 “rule learning heuristic should (among other aspects) take both, the consistency and coverage of a rule, into account… recall focuses on coverage of a rule. It measures the fraction of covered instances” Eq.4 δrec(C):=TP/(TP+FN) specifies coverage rate/fraction using true positive and false negative, instances are of the underlying data, e.g. [P.14 ¶2] “coverage should be fine-tuned depending on the data set at hand”. Further, [P.6-7 Sect3.3] threshold selection includes maximizing coverage conveys a preset proportion}.
Rapp is directed to machine learning with heuristics thus being analogous. A person having ordinary skill in the art would have considered it obvious prior to the effective filing date to specify coverage per Rapp to support Cha’s coverage more clearly to arrive at the invention as claimed for the stated motivation “Motivation and goals… we argue that all common rule learning heuristics basically trade off between two aspects, consistency and coverage. Our long-term goal is to better understand how these two aspects should be weighed to assess the quality of candidate rules during training” [P.2 ¶2-3] and because “rules with great coverage, i.e., rules that cover a large number of instances, tend to be more reliable” [P.4 Sect2.4].
With respect to claim 13, which recites adopting the method according to claim 5 as well as the full body of claim 8 system, the system is taught by the combination of Gooch, Cha and Bar, and further claim 5 is taught in combination with Rapp. Therefore, the rejection of claims 8 and 5 with equal motivation are applied to claim 13.
With respect to claim 19, which recites adopting the method according to claim 5 as well as the full body of claim 9 device, the device is taught by the combination of Gooch, Cha and Bar, and further claim 5 is taught in combination with Rapp. Therefore, the rejection of claims 9 and 5 with equal motivation are applied to claim 19.
Claims 6-7, 14-15 and 20-21 are rejected under 35 U.S.C. 103 as being unpatentable over Gooch, Cha and Bar in view of:
Zhao et al., “Generalizable Meta-Heuristic based on Temporal Estimation of Rewards for Large Scale Blackbox Optimization” hereinafter Zhao (arXiv: 1812.06585v2).
With respect to claim 6, the combination of Gooch, Cha and Bar teaches the machine heuristic learning method for operation behavior record management according to claim 1, wherein
the evaluation data generated from the basic working condition data and the operation data comprises an optimization objective value or a restrictive result value {Zhao [P.2] Def.1 Best Fitness “evaluation, the best known objective value (minimum for minimization or maximum for maximization) queried via function evaluation from the start of the optimization process until t is called best fitness” e.g. Def.4 “where y*(A, y0, T) is the best fitness” is an optimization objective value. See also per [P.4] Alg.1 Output: best}.
Zhao is directed to machine learning with heuristics thus being analogous. A person having ordinary skill in the art would have considered it obvious prior to the effective filing date to perform evaluation by best fitness per Zhao in combination to arrive at the invention as claimed for the motivation of performance benchmarking/baselines [P.5 Tbl.1] by using “principled guidelines for practical use” [P.2 ¶2] and/or to show “performance rankings are statistically meaningful” [P.5 ¶2].
With respect to claim 7, the combination of Gooch, Cha, Bar and Zhao teaches the machine heuristic learning method for operation behavior record management according to claim 6, wherein
if the evaluation data is superior to recorded evaluation data corresponding to other operation data under the same basic working condition data, an operation behavior record set is updated {Zhao [P.4] Alg.1 Lines11-12 “update xbest, ybest” and “Q.add(<xbest, ybest, Δt>); //add record” where best is based on Best Fitness [P.2] Def.1. Further, ranking performance establishes superior Tables 1,4-5.}.
A person having ordinary skill in the art would have considered it obvious prior to the effective filing date to update a record set with superior/best data per Zhao to arrive at the invention as claimed for a motivation as in claim 6 and further “we should improve the best fitness by the most” [P.2 Last¶] as well as “for practical use and simplicity to ensure easy generalization” [P.7 Last¶].
With respect to claim 14, which recites adopting the method according to claim 6 as well as the full body of claim 8 system, the system is taught by the combination of Gooch, Cha and Bar, and further claim 6 is taught in combination with Zhao. Therefore, the rejection of claims 8 and 6 with equal motivation are applied to claim 14.
With respect to claim 15, which recites adopting the method according to claim 7 as well as the full body of claim 8 system, the system is taught by the combination of Gooch, Cha and Bar, and further claim 7 is taught in combination with Zhao. Therefore, the rejection of claims 8 and 7 with equal motivation are applied to claim 15.
With respect to claim 20, which recites adopting the method according to claim 6 as well as the full body of claim 9 device, the device is taught by the combination of Gooch, Cha and Bar, and further claim 6 is taught in combination with Zhao. Therefore, the rejection of claims 9 and 6 with equal motivation are applied to claim 20.
With respect to claim 21, which recites adopting the method according to claim 7 as well as the full body of claim 9 device, the device is taught by the combination of Gooch, Cha and Bar, and further claim 7 is taught in combination with Zhao. Therefore, the rejection of claims 9 and 7 with equal motivation are applied to claim 21.
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure:
Both of Gooch and Bar have Chinese equivalent in patent family if translation aids applicant, see Gooch: CN113039495A, and Bar: CN114430815A1
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to Chase P Hinckley whose telephone number is (571)272-7935. The examiner can normally be reached M-F 9:00 - 5:00.
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, Miranda M. Huang can be reached at 571-270-7092. 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.
/CHASE P. HINCKLEY/Examiner, Art Unit 2124