DETAILED ACTION
This office action is in response to claims filed 1 June 2026.
Claims 1-10, 12-23, and 25-32 are pending.
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 .
Response to Arguments
Applicant's arguments filed 1 June 2026 have been fully considered but they are not persuasive.
On pages 12-14, the applicant argues in the remarks:
“Applicant does not contest the Examiner’s interpretation of these claims [27 and 32] under 35 U.S.C. 112(f)…
“Claims 14 and 25 have been amended to recite sufficient structure for each identified information, thereby avoiding interpretation under 35 U.S.C. 112 (f).
“Claim 14 as amended now recites structural terms for each identified limitation…The term ‘computing systems’ is not a generic placeholder but rather a structural term that connotes sufficiently definite structure to one of ordinary skill in the art. See MPEP 2181, subsection I.A…the claim language recites sufficient structure to avoid interpretation under 35 U.S.C. 112(f). Alternatively, even if 35 U.S.C. 112(f) were to apply, the specification provides adequate structural support for these limitations…
Claim 25 as amended now recites a learning engine comprising an On-Chip Learning Engine with Reward Calculator, Weight Updated Unit, Backpropagation Engine, and Feedback Buffer. (see paragraphs [0152]-[0156].
“The amendments to claim 14 and 25 add specific structural components that perform the recited functions, thereby satisfying prong C of the three-prong test for 35 U.S.C. 112(f)…”
The examiner respectfully disagrees. With the exception of “central executive core” now reciting a processing core and therefore reciting sufficient structure to avoid an interpretation under 35 U.S.C. 112(f), none of the amendments are sufficient to avoid an interpretation under 35 U.S.C. 112(f). For example, regarding the experiential and analytical system, the term “one or more computing systems” is a generic placeholder and is not a structural term that connotes sufficiently definite structure, as the applicant alleges. For example, the term “computing system” does not describe any underlying hardware or software, like computers, processors, cores, or memory. A “computing system” could describe a system comprising structure that is only hardware, only software, or a combination of both. As such, “computing systems” are generic placeholders.
Furthermore, the amendments to the claim did not address “each identified limitation”, as the applicant alleges. For example, the identified limitations include “integration and validation unit”, “interface controller”, and “learning engine”. None of these limitations were modified via amendment to claim 14, and the applicant’s argument does not specifically address them.
Whether the specification provides adequate structural support for claim limitations or not is irrelevant to whether or not those claim limitations should be interpreted under 35 U.S.C. 112(f), because in prong 3, the sufficient structure must be in the claim, not the specification. Since none of those details are in the claim, they are irrelevant.
The interpretations in the current office action are therefore maintained, and the applicant’s argument is not persuasive.
On pages 14-17, the applicant argues in the remarks:
“Claim 1 as amended recites an experiential system comprising computing systems executing LLMs and an analytical system comprising computing systems executing physical world models. These are not mental processes…Claims do not recite a mental process when they do not contain limitations that can practically be performed in the human mind, for instance when the human mind is not equipped to perform the claim limitations.
The examiner respectfully notes that 1) the claim does not actually recite execution of LLMs and analytical systems. At best, the claim merely makes a decision to route user input to one of the LLM or the analytical system, and receiving outputs from the LLM or the analytical system. Since the claim does not recite execution of either the experiential system or the analytical system, the applicant’s argument is not persuasive. 2) If the claim had recited execution of LLMs and analytical systems (which it does not), the examiner agrees that this hypothetical recitation would not fall into the mental process grouping. However, such execution of a machine learning LLM, or an analytical computing system would still fail to provide eligibility, because they represent additional elements that fail to integrate the judicial exception into a practical application, or provide significantly more. Specifically, the previous rejection set forth that the claim recited a judicial exception (mental processes of performing routing analysis, making a routing decision, and generating a final result), and a hypothetical recitation of execution of a large language model, or analytical computing system represents mere application or implementation of the abstract idea on or using a computer tool, where the computer tool is the machine learning LLM, or the analytical computation model. Since the hypothetical recitation would not provide eligibility, the applicant’s argument is not persuasive.
On pages 14-17, the applicant argues in the remarks:
“The routing decision is not a mental process when considered in the context of the claim as a whole. Claim 1 requires routing user input to computing systems that execute LLMs or physical world models, a technical operation directing computational tasks to specialized processing subsystems…This is an automated computational process allocating resources between heterogeneous processing systems, not mental judgment.”
The examiner respectfully disagrees. Nothing in the claim language, or applicant’s argument, precludes the routing decision from being a decision able to be performed completely in the human mind. The claim contains no detail about how the routing decision is made other than to say it is “based on routing analysis”, and that the routing analysis is itself performed “on the basis of a plurality of characteristics.” However, the claims do not describe these characteristics, let alone in a non-generic manner that would preclude the analysis from being performed in the human mind. Further, the claims do not describe how the decision is made based on the analysis, let alone in a non-generic manner that would preclude the decision from being performed in the human mind. For example, based on a general analysis of a content of a user input, or desired output, it is well within the capabilities of the human mind to decide whether to route that input to a particular type of machine learning system, like a LLM or analytical system. Therefore, the applicant’s argument is not persuasive.
On pages 14-17, the applicant argues in the remarks:
“As recited in new dependent claims 30-32, the validation triggers concrete remedial actions when outputs fail. Re-routing to a different processing pathway, iteratively refining outputs until physical constraints are satisfied, or flagging uncertainties and presenting alternative outputs. This is analogous to the remedial network security actions…found to integrate the abstract idea into a practical application in USPTO Subject Matter Eligibility Example 47…Accordingly, the claims as amended do not recite a mental process under Step 2A, Prong 1.”
The examiner respectfully disagrees. Initially, the examiner points out that whether a claim recites integration into a practical application has no bearing on whether a claim “recites a mental process under Step 2A prong 1” (i.e., a claim reciting a mental process under Step 2A prong 1, may potentially have additional elements that can either integrate that judicial exception into a practical application or not. This is analyzed under Step 2A prong 2, not prong 1). As for the elements of claims 30-32, they are analyzed under step 2A prong 2 and step 2B below, but are also briefly discussed here: re-routing to a different processing pathway is a mental process, akin to making the initial routing decision. Iteratively refining outputs until constraints are satisfied represents mere application of the LLM or analytical system as a tool to implement the judicial exception, flagging uncertainties is a mental process, and presenting alternative outputs is mere data output that is insignificant extra-solution activity, as well as well-understood, routine, and conventional activity. Since none of these alternative limitations provide eligibility, the claim as a whole is ineligible, and the applicant’s argument is not persuasive.
On pages 14-17, the applicant argues in the remarks:
“The claims provide a specific technical improvement to AI systems. The invention addresses the ‘flat world’ limitation of LLMs, which have a two-dimensional understanding confined to linguistic patters without intrinsic knowledge of physical reality. See As-Filed Specification, paragraph [0035]. The hybrid architecture augments LLMs with physical world models providing the ‘world map’ they lack…This is a specific technical improvement to AI system functioning, not merely applying an abstract idea to a computer.”
The examiner respectfully disagrees. The claim does not improve AI or LLM systems as alleged, because the claim simply decides whether to route input to a LLM or a separate analytical system. The LLM is not augmented with the analytical system, or a “physical world model”, because the LLM is not changed or modified at all. Therefore the applicant’s argument is not persuasive.
On pages 17-22, the applicant argues that the combination of Howard and Penfield, fails to teach the newly amended elements of the independent claims, and that the dependent claims are allowable for their dependency. These arguments are moot because they do not specifically challenge the new references (LEE, ARORA, and XU, cited below) used in the current rejection.
Examiner’s Note
Claim 30 that was previously presented on 22 October 2025 appears to have been replaced by a new version of claim 30 filed 1 June 2026, without an indication that the claim has been canceled or amended. Applicant is reminded of the requirement of 37 CFR 1.121(I)(C)(A): “The current status of all of the claims in the application, including any previously canceled or withdrawn claims, must be given. Status is indicated in a parenthetical expression following the claim number by one of the following status identifiers: (original), (currently amended), (previously presented), (canceled), (withdrawn), (new), or (not entered).” For examination purposes, the examiner will interpret the previously presented claim 30 as being canceled, and will examine the new claim 30.
Additionally, at least in claim 1, the claim contains a limitation with words that are both underlined, and struck through, specifically “models and physical data for”. For examination purposes, the examiner will interpret the struck through words as being canceled.
Claim Interpretation
The following is a quotation of 35 U.S.C. 112(f):
(f) Element in Claim for a Combination. – An element in a claim for a combination may be expressed as a means or step for performing a specified function without the recital of structure, material, or acts in support thereof, and such claim shall be construed to cover the corresponding structure, material, or acts described in the specification and equivalents thereof.
The following is a quotation of pre-AIA 35 U.S.C. 112, sixth paragraph:
An element in a claim for a combination may be expressed as a means or step for performing a specified function without the recital of structure, material, or acts in support thereof, and such claim shall be construed to cover the corresponding structure, material, or acts described in the specification and equivalents thereof.
The claims in this application are given their broadest reasonable interpretation using the plain meaning of the claim language in light of the specification as it would be understood by one of ordinary skill in the art. The broadest reasonable interpretation of a claim element (also commonly referred to as a claim limitation) is limited by the description in the specification when 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, is invoked.
As explained in MPEP § 2181, subsection I, claim limitations that meet the following three-prong test will be interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph:
(A) the claim limitation uses the term “means” or “step” or a term used as a substitute for “means” that is a generic placeholder (also called a nonce term or a non-structural term having no specific structural meaning) for performing the claimed function;
(B) the term “means” or “step” or the generic placeholder is modified by functional language, typically, but not always linked by the transition word “for” (e.g., “means for”) or another linking word or phrase, such as “configured to” or “so that”; and
(C) the term “means” or “step” or the generic placeholder is not modified by sufficient structure, material, or acts for performing the claimed function.
Use of the word “means” (or “step”) in a claim with functional language creates a rebuttable presumption that the claim limitation is to be treated in accordance with 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph. The presumption that the claim limitation is interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, is rebutted when the claim limitation recites sufficient structure, material, or acts to entirely perform the recited function.
Absence of the word “means” (or “step”) in a claim creates a rebuttable presumption that the claim limitation is not to be treated in accordance with 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph. The presumption that the claim limitation is not interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, is rebutted when the claim limitation recites function without reciting sufficient structure, material or acts to entirely perform the recited function.
Claim limitations in this application that use the word “means” (or “step”) are being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, except as otherwise indicated in an Office action. Conversely, claim limitations in this application that do not use the word “means” (or “step”) are not being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, except as otherwise indicated in an Office action. The limitations using the words “means” are present in claim 27.
This application includes one or more claim limitations that do not use the word “means,” but are nonetheless being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, because the claim limitation(s) uses a generic placeholder that is coupled with functional language without reciting sufficient structure to perform the recited function and the generic placeholder is not preceded by a structural modifier. Such claim limitation(s) are: “experiential system”, “analytical system”, “integration and validation unit”, “interface controller”, and “learning engine” in claims 14, 18-23 and 25.
Because this/these claim limitation(s) is/are being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, it/they is/are being interpreted to cover the corresponding structure described in the specification as performing the claimed function, and equivalents thereof, recited as “system-on-chip” architecture comprising hardware that executes the software elements at issue.
If applicant does not intend to have this/these limitation(s) interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, applicant may: (1) amend the claim limitation(s) to avoid it/them being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph (e.g., by reciting sufficient structure to perform the claimed function); or (2) present a sufficient showing that the claim limitation(s) recite(s) sufficient structure to perform the claimed function so as to avoid it/them being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph.
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 1-10, 12-23, and 25-32 are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea (mental process) without significantly more.
Regarding claim 1, in step 1 of the 101 analysis set forth in MPEP 2106, the claim recites a method that makes routing decisions between experiential systems and analytical systems based on input for AI user requests. A method is one of the four statutory categories of invention.
In step 2A, prong 1 of the 101 analysis set forth in the MPEP 2106, the examiner has determined that the following limitations recite a process that, under the broadest reasonable interpretation, covers a mental process but for recitation of generic computer components:
i. “performing a routing analysis on the user input on a basis of a plurality of characteristics” (a person can mentally perform routing analysis by simply evaluating characteristics to mentally analyze them (MPEP 2106.04(a)))
ii. “making a routing decision based on the routing analysis to send the user input to one or both of: an experiential system; and an analytical system” (a person can mentally make a routing decision by simply evaluating the routing analysis and making a judgement of where to route user input (MPEP 2106.04(a)))
iii. “generating a final result by performing a result validation procedure on the one or more outputs” (a person can mentally generate results by evaluating validation criteria and making a judgement that an output is valid (MPEP 2106.04(a))).
If claim limitations, under their broadest reasonable interpretation, covers performance of the limitations as a mental process but for the recitation of generic computer components, then it falls within the mental process grouping of abstract ideas. Accordingly, the claim “recites” an abstract idea.
In step 2A, prong 2 of the 101 analysis set forth in MPEP 2106, the examiner has determined that the following additional elements do not integrate this judicial exception into a practical application:
iv. “a method for processing artificial intelligence (AI) user requests using a system that comprises both an experiential system and an analytical system” (mere instructions to implement an abstract idea on a computer, or merely uses a computer as a tool to perform an abstract idea (MPEP 2106.05(f))).
v. “receiving a user input from a user device” (insignificant extra-solution activity of mere data gathering (MPEP 2106.05(g))).
vi. “an experiential system comprising one or more computing system configured to execute one or more large language models implementing generative AI systems” (generally links the use of the judicial exception to a particular technological environment or field of use (MPEP 2106.05(h))).
vii. “an analytical system comprising one or more computing systems configured to execute physical world models that implement one or more physical constraints derived from physical laws governing physical real-world phenomena and processes” (generally links the use of the judicial exception to a particular technological environment or field of use (MPEP 2106.05(h))).
viii. “routing the user input to at least one of the experiential system and the analytical system responsive to the routing decision” (insignificant extra-solution activity of mere data output (MPEP 2106.05(g))).
ix. “receiving one or more outputs from at least one of the experiential system and the analytical system” (insignificant extra-solution activity of mere data gathering (MPEP 2106.05(g))).
x. “wherein the result validation procedure validates the one or more outputs against the one or more physical constraints” (insignificant extra-solution activity of mere data output (MPEP 2106.05(g))).
xi. “transmitting the final result to the user device” (insignificant extra-solution activity of mere data output (MPEP 2106.05(g))).
Since the claim does not contain any other additional elements that are indicative of integration into a practical application, the claim is “directed” to an abstract idea.
In step 2B of the 101 analysis set forth in the 2019 PEG, the examiner has determined through reanalysis of the following limitations considered in step 2A prong 2, that the claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception.
iv. “a method for processing artificial intelligence (AI) user requests using a system that comprises both an experiential system and an analytical system” (mere instructions to implement an abstract idea on a computer, or merely uses a computer as a tool to perform an abstract idea (MPEP 2106.05(f))).
v. “receiving a user input from a user device” (well-understood, routine, and conventional activity of receiving data over a network, (MPEP 2106.05(d)(II))).
vi. “an experiential system comprising one or more computing system configured to execute one or more large language models implementing generative AI systems” (generally links the use of the judicial exception to a particular technological environment or field of use (MPEP 2106.05(h))).
vii. “an analytical system comprising one or more computing systems configured top execute physical world models that implement one or more physical constraints derived from physical laws governing physical real-world phenomena and processes” (generally links the use of the judicial exception to a particular technological environment or field of use (MPEP 2106.05(h))).
viii. “routing the user input to at least one of the experiential system and the analytical system responsive to the routing decision” (well-understood, routine, and conventional activity of transmitting data over a network, (MPEP 2106.05(d)(II))).
ix. “receiving one or more outputs from at least one of the experiential system and the analytical system” (well-understood, routine, and conventional activity of receiving data over a network, (MPEP 2106.05(d)(II))).
x. “wherein the result validation procedure validates the one or more outputs against the one or more physical constraints” (insignificant extra-solution activity of mere data output (MPEP 2106.05(g))).
xi. “transmitting the final result to the user device” (well-understood, routine, and conventional activity of transmitting data over a network, (MPEP 2106.05(d)(II))).
Considering the additional elements individually and in combination, and the claim as a whole, the additional elements do not provide significantly more than the abstract idea. Therefore, the claim is not patent eligible.
Regarding claim 2, the additional element “the plurality of characteristics comprises at least two of: task type classification; complexity assessment; domain identification; temporal analysis; risk evaluation; and resource estimation” does not render the claim patent eligible because under step 2A prong 2, it does not integrate the judicial exception into a practical application (generally links the use of the judicial exception to a particular technological environment or field of use (MPEP 2106.05(h)), and under step 2B it does not amount to significantly more than the judicial exception (generally links the use of the judicial exception to a particular technological environment or field of use (MPEP 2106.05(h)).
Regarding claim 3, the additional element “the routing analysis is optimized by at least one of: reinforcement learning; supervised learning; or a combination of reinforcement learning and supervised learning” does not render the claim patent eligible because under step 2A prong 2, it does not integrate the judicial exception into a practical application (generally links the use of the judicial exception to a particular technological environment or field of use (MPEP 2106.05(h)), and under step 2B it does not amount to significantly more than the judicial exception (generally links the use of the judicial exception to a particular technological environment or field of use (MPEP 2106.05(h)).
Regarding claim 4, the additional element “the routing analysis is performed in at least one of: a performance mode configured to prioritize response time; an accuracy mode configured to prioritize accuracy; an efficiency mode configured to minimize at least one of computational cost and energy; a safety mode configured to maximize validation rigor; and a balanced mode configured to optimize weighted combination of multiple objectives” does not render the claim patent eligible because under step 2A prong 2, it does not integrate the judicial exception into a practical application (generally links the use of the judicial exception to a particular technological environment or field of use (MPEP 2106.05(h)), and under step 2B it does not amount to significantly more than the judicial exception (generally links the use of the judicial exception to a particular technological environment or field of use (MPEP 2106.05(h)).
Regarding claim 5, the additional element “the experiential system is configured to operate without explicit modeling of physical laws, mathematical constraints, or causal mechanisms” does not render the claim patent eligible because under step 2A prong 2, it does not integrate the judicial exception into a practical application (generally links the use of the judicial exception to a particular technological environment or field of use (MPEP 2106.05(h)), and under step 2B it does not amount to significantly more than the judicial exception (generally links the use of the judicial exception to a particular technological environment or field of use (MPEP 2106.05(h)).
Regarding claim 6, the additional element “the analytical system comprises two or more computational implementations of scientific principles directed to: physics; chemistry; biology; economics; and engineering” does not render the claim patent eligible because under step 2A prong 2, it does not integrate the judicial exception into a practical application (generally links the use of the judicial exception to a particular technological environment or field of use (MPEP 2106.05(h)), and under step 2B it does not amount to significantly more than the judicial exception (generally links the use of the judicial exception to a particular technological environment or field of use (MPEP 2106.05(h)).
Regarding claim 7, the additional element “the analytical system is operable to generate an output comprising at least one of: an uncertainty quantification; a sensitivity analysis; a validation certificate; a documenting constraint satisfaction; and traceability information” does not render the claim patent eligible because under step 2A prong 2, it does not integrate the judicial exception into a practical application (generally links the use of the judicial exception to a particular technological environment or field of use (MPEP 2106.05(h)), and under step 2B it does not amount to significantly more than the judicial exception (generally links the use of the judicial exception to a particular technological environment or field of use (MPEP 2106.05(h)).
Regarding claim 8, the additional element “the result validation procedure comprises implementing one or more consistency checking algorithms directed to: numerical consistency; logical consistency; semantic consistency; and physical consistency” does not render the claim patent eligible because under step 2A prong 1, it recites a judicial exception (mental process) (a person can mentally validate output by simply evaluating an output and making a judgement that the output is consistent either numerically, logically, semantically, or physically (MPEP 2106.04(a))).
Regarding claim 9, the additional element “the result validation procedure is operable to compute one or more confidence metrics comprising at least one of: model agreement; constraint satisfaction margins; historical accuracy; uncertainty quantification; and validation results. ” does not render the claim patent eligible because under step 2A prong 2, it does not integrate the judicial exception into a practical application (generally links the use of the judicial exception to a particular technological environment or field of use (MPEP 2106.05(h)), and under step 2B it does not amount to significantly more than the judicial exception (generally links the use of the judicial exception to a particular technological environment or field of use (MPEP 2106.05(h)).
Regarding claim 10, the additional element “the result validation procedure is operable to merge a first output received from the experiential system comprised by the one or more outputs and a second output received from the analytical system comprises by the one or more outputs” does not render the claim patent eligible because under step 2A prong 2, it does not integrate the judicial exception into a practical application (generally links the use of the judicial exception to a particular technological environment or field of use (MPEP 2106.05(h)), and under step 2B it does not amount to significantly more than the judicial exception (generally links the use of the judicial exception to a particular technological environment or field of use (MPEP 2106.05(h)).
Regarding claim 12, the additional element “executing a feedback procedure comprising at least one of: adjusting one or more parameters for performing the routing analysis; updating the experiential system; and updating the analytical system” does not render the claim patent eligible because under step 2A prong 1, it recites a judicial exception (mental process) (a person can mentally adjust parameters by simply making an evaluation of parameters for performing the routing analysis, and making a judgement of adjusted parameters(MPEP 2106.04(a))).
Regarding claim 13, the additional element “the feedback procedure is performed responsive to at least one of: outcomes; performance metrics; and user feedback associated with the final result. ” does not render the claim patent eligible because under step 2A prong 2, it does not integrate the judicial exception into a practical application (generally links the use of the judicial exception to a particular technological environment or field of use (MPEP 2106.05(h)), and under step 2B it does not amount to significantly more than the judicial exception (generally links the use of the judicial exception to a particular technological environment or field of use (MPEP 2106.05(h)).
Regarding claim 30, the additional element “responsive to the result validation procedure determining that the one or more outputs fail to satisfy the one or more physical constraints ” does not render the claim patent eligible because under step 2A prong 1, it recites a judicial exception (mental process) (a person can mentally determine whether a constraint is satisfied by simply evaluating the constraint and making a judgment that it has been satisfied or not (MPEP 2106.04(a))). Further, the additional element “the method further comprises performing one or more of: re-routing the user input to a different processing pathway; iteratively refining the one or more outputs until the one or more physical constraints are satisfied; and flagging uncertainties and presenting alternative outputs” does not render the claim patent eligible because under step 2A prong 1, it recites a judicial exception (mental process) (a person can mentally reroute user input by simply evaluating processing pathways and making a judgement to route input to a different pathway, or mentally flag uncertainties by simply evaluating uncertainties and making a judgement of emphasizing certain uncertainties (MPEP 2106.04(a))). Further, this additional limitation does not render the claim patent eligible because under step 2A prong 2, it does not integrate the judicial exception into a practical application (iteratively refining input represents merely adding the words “apply it” (or an equivalent) with the judicial exception, or mere instructions to implement an abstract idea on a computer, or merely uses a computer as a tool to perform an abstract idea (MPEP 2106.05(f)), and under step 2B it does not amount to significantly more than the judicial exception (iteratively refining input represents merely adding the words “apply it” (or an equivalent) with the judicial exception, or mere instructions to implement an abstract idea on a computer, or merely uses a computer as a tool to perform an abstract idea (MPEP 2106.05(f)). Further, this additional limitation does not render the claim patent eligible because under step 2A prong 2, it does not integrate the judicial exception into a practical application (presenting output represents insignificant extra-solution activity of mere data output (MPEP 2106.05(g))), and under step 2B it does not amount to significantly more than the judicial exception (presenting output represents well-understood, routine, and conventional activity of transmitting data over a network, (MPEP 2106.05(d)(II))).
Regarding claims 14-23, 25-29, and 31-32, they comprise limitations similar to those of claims 1-10, and 12-13, and are therefore rejected for similar rationale.
Claim Rejections - 35 USC § 103
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.
Claims 1-5, 9-10, 12-18, 22-23, and 25-30 are rejected under 35 U.S.C. 103 as being unpatentable over HOWARD Pub. No.: US 2021/0334679 A1 (hereafter HOWARD), in view of LEE, “What Are Large Language Models Used For?” (available at https://blogs.nvidia.com/blog/what-are-large-language-models-used-for/ on 26 January 2023, hereafter LEE), in view of ARORA et al. “Bayesian Networks for Risk Prediction Using Real World Data: A Tool for Precision Medicine.” (available 2019, hereafter ARORA) in view of PENFIELD et al. Patent No.: US 11,893,048 B1 (hereafter PENFIELD), in view of XU et al. Pub. No.: US 2024/0086767 A1 (hereafter XU).
HOWARD, and PENFIELD were cited previously
Regarding claim 1, HOWARD teaches the invention substantially as claimed, including:
A method for processing artificial intelligence (AI) user requests using a system that comprises both an experiential system and an analytical system, comprising:
receiving a user input from a user device ([0065] An exemplary block diagram of a system 100 according to the present techniques is shown in FIG. 1. System 100 may include, for example, three layers, Input Data Layer 102…Input Data Layer 102 may include data-capturing points from data channels 108 associated with types of data: video, image, text, audio, etc., as well as meta world data 110 and objective data 112. The data channels layer may include several stages of data retrieval and manipulation, such as: identification of input points and types for each data channel, retrieval of data and data preprocessing, and data sampling techniques and storage);
performing a routing analysis on the user input on a basis of a plurality of characteristics; making a routing decision based on the routing analysis ([0057] Embodiments may provide an intelligent adaptive system that combines input data types, processing history and objectives, research knowledge and situational context to determine what is the most appropriate mathematical model. [0066] Model selector 114 identify a set of methods and operations from model repository 116 to apply on the input data in relation to intelligence inferring and pattern determination. Such mechanisms may include the stages such as a Critic-Selector Mechanism, which may be based on combining input data types from data channels 108, meta world data 110, such as processing history, and objective data 112, including research knowledge and situational context to determine what is the most appropriate Artificial Intelligence (AI) model for existing data and how the system should manage the processing resources, be it models or computing infrastructure (i.e., determining the most appropriate mathematical model represents making a “routing decision” based on analysis of the input data and a plurality of characteristics including data types, processing history and objectives, etc.)) to send the user input to one or both of:
an experiential system ([0121] It is to be noted that any type of machine learning model may be utilized by Selector Component 448 for selection of models, as well as generation of models. For example, as shown in FIG. 8a , embodiments may utilize…deep learning models 819, such as random, recurrent, and recursive neural network models (RNNs) 820 (i.e., at least recurrent neural network models represent systems that implement experiential models, according to [0173] of the specification))…and
an analytical system [0121] It is to be noted that any type of machine learning model may be utilized by Selector Component 448 for selection of models, as well as generation of models. For example, as shown in FIG. 8a , embodiments may utilize…Bayesian learning models 811, such as sparse Bayes models 812, naive Bayes models 813, and expectation maximization models 814 (i.e., at least Bayesian learning models represent systems that implement mathematical analytical models, according to [0181] and [0182] of the specification. HOWARD also discusses selection of other analytical systems including linear regression models))…
routing the user input to at least one of the experiential system and the analytical system responsive to the routing decision ([0108] Solution Processor 456. Solution Processor 456 may receive the scheduled tasks or process modules 457 and runs them, if needed in parallel, on the appropriate computing infrastructure (i.e., tasks representing the input are sent to, or “routed” to the selected models executing on the appropriate computing infrastructure));
receiving one or more outputs from at least one of the experiential system and the analytical system ([0067] Output Data Layer 106 may include the results of running the resulting model or ensemble of models on the automatically selected computing infrastructure)…
transmitting the final result to the user device ([0202] Agent layer 2502 may include digital hardware and software 2526 to provide system input and output to users.).
While HOWARD discusses implementing an experiential deep learning model such as neural networks, HOWARD does not explicitly teach:
an experiential system comprising one or more computing system configured to execute one or more large language models implementing generative AI systems;
However, in analogous art that similarly discusses examples of deep learning models, LEE teaches:
an experiential system comprising one or more computing system configured to execute one or more large language models implementing generative AI systems (AI applications are summarizing articles, writing stories and engaging in long conversations — and large language models are doing the heavy lifting. A large language model, or LLM, is a deep learning algorithm that can recognize, summarize, translate, predict and generate text and other forms of content based on knowledge gained from massive datasets (i.e., large language models are implemented as examples of deep learning algorithms used to process and generate specific types of language related data)).
It would have been obvious to a person having ordinary skill in the art before the effective filing date of the invention to have combined LEE’s teaching of a large language model as a deep learning algorithm, with HOWARD’s teaching of implementing a deep learning algorithm to process user input, to realize, with a reasonable expectation of success, a system that implements a deep learning algorithm to process input, as in HOWARD, where the algorithm is a LLM due to the input being language related, as in LEE. A person having ordinary skill would have been motivated to make this combination to enable a deep learning model to process huge volumes of text to better generate responses in sentences or paragraphs (LEE)
While HOWARD discusses implementing a Bayesian learning model as an analytical system, HOWARD and LEE do not explicitly teach:
an analytical system comprising one or more computing systems configured to execute physical world models that implement one or more physical constraints derived from physical laws governing physical real-world phenomena and processes
However, in analogous art that similarly discusses Bayesian learning models, ARORA teaches:
an analytical system comprising one or more computing systems configured to execute physical world models that implement one or more physical constraints derived from physical laws governing physical real-world phenomena and processes ([Pages 440-441] There are three approaches to the construction of BN [Bayesian network] structures: purely expert-elicited, purely automated or machine-learned, and a combined approach where prior expert knowledge is incorporated into the automated learning process…Purely machine-learned approaches include constraint- and score-based learning. Constraint-based approaches use conditional interdependencies in the data to derive the model structure…The combined approach allows expert or user input to force known temporal relationships, direct relationships, and direct causal relationships to be part of a machine-learned proposed structure (i.e., the proposed combined approach results in Bayesian networks that model predictions of real-world risk based at least partially on constraints in real-world data that are forced by experts or users))
It would have been obvious to a person having ordinary skill in the art before the effective filing date of the invention to have combined ARORA’s teaching of a Bayesian network that models predictions of real world risk based at least partially on constraints in real-world data that are forced by experts or users, with HOWARD and LEE’s teaching of implementing Bayesian networks to process user input, to realize, with a reasonable expectation of success, a system that implements Bayesian networks to process user input, as in HOWARD and LEE, to implement a real world risk prediction based at least partially on real-world constraints, as in ARORA. A person having ordinary skill in the art would have been motivated this combination to improve scalability and compactness of networks (ARORA Page 440).
While HOWARD discusses receiving results from selected models and transmitting output to users, HOWARD, LEE, and ARORA does not explicitly teach:
generating a final result by performing a result validation procedure on the one or more outputs
However, in analogous art that similarly discusses receiving results from machine learning models, PENFIELD teaches:
receiving one or more outputs from at least one of the experiential system and the analytical system; generating a final result by performing a result validation procedure on the one or more outputs ([Column 18, Line 66-Column 19, Line 8] The output 1235 produced by each pre-trained machine learning model can include individual inferences for each field. Multiple outputs, or inferences may be produced or output 1235 by each network, whereby some are relevant and others may not be. Verification and validation checks can also be applied to either select from candidate outputs or verify and validate the accuracy or relevance of these outputs. After all validation checks are passed, the indexes (or model inferences) from each field are combined and returned and saved in the database 1240).
It would have been obvious to a person having ordinary skill in the art before the effective filing date of the invention to have combined PENFIELD’s teaching of validating outputs from machine learning models, with HOWARD’s teaching of generating outputs from experiential or analytical machine learning models, to realize, with a reasonable expectation of success, a system that generates outputs from experiential or analytical machine learning models, as in HOWARD, and validates the outputs before combining them, as in PENFIELD. A person having ordinary skill would have been motivated to make this combination to ensure accuracy and relevance of outputs (PENFIELD [Column 18, Line 66-Column 19, Line 8]).
While PENFIELD discusses generating results based on a validation, HOWARD, LEE, ARORA, and PENFIELD do not explicitly teach:
wherein the result validation procedure validates the one or more outputs against the one or more physical constraints;
However, in analogous art that similarly teaches generating and validating results, XU teaches:
wherein the result validation procedure validates the one or more outputs against the one or more physical constraints ([0153] the hyperparameter tuner 550 trains/validates the machine learning model by tuning the set of hyperparameters to achieve an optimal performance with respect to the different weighted metrics. In some instances, this process is performed while ensuring that each of the one or more constraints is satisfied…Upon optimizing the machine learning model for the plurality of metrics, the hyperparameter tuning system 500 outputs the trained/validated ML model along with configuration of the hyperparameters that optimize the objective function, and in some instances satisfy the one or more constraints (i.e., results of a trained machine learning model are validated to ensure that physical constraints are satisfied));
It would have been obvious to a person having ordinary skill in the art before the effective filing date of the invention to have combined XU’s teaching of validating machine learning model outputs with constraints, with the combination of HOWARD, LEE, ARORA, and PENFIELD’s teaching of selecting and validating an experiential large language model or an analytical Bayesian model, to realize, with a reasonable expectation of success, a system that validates a selected experiential or analytical model, as in HOWARD, LEE, ARORA, and PENFIELD, against one or more physical constraints, as in XU. A person having ordinary skill would have been motivated to make this combination to enable an optimal machine learning model to be selected that conforms to specified constraints.
Regarding claim 2, HOWARD further teaches:
the plurality of characteristics comprises at least two of: task type classification; complexity assessment; domain identification; temporal analysis; risk evaluation; and resource estimation ([0070] The critic-selector mechanism may process the problem description, recognize the problem type (i.e., “task type classification”), and then activate the selector component. The selector may start up several sets of resources (models or combination of models), which were learned from experience as the most probable viable approaches (i.e., “estimating” probable “resources” for a type of problem or task) for the given situation at hand).
Regarding claim 3, HOWARD further teaches:
the routing analysis is optimized by at least one of: reinforcement learning; supervised learning; or a combination of reinforcement learning and supervised learning ([0118] Orchestrator Perspective. From a more abstract, higher level point of view, system 400 may be seen as an orchestrator-centered system 800 managing all possible types of models, which may be organized in a graph, and which can be used for selecting processing paths, as illustrated in FIGS. 8a-c . Orchestrator 800 may use any approach from logic and planning, supervised to unsupervised learning, reinforcement learning, search algorithms, or any combination of those (i.e., orchestrator utilizes at least reinforcement learning to search for, and select a type of model)).
Regarding claim 4, HOWARD further teaches:
the routing analysis is performed in at least one of:
a performance mode configured to prioritize response time ([0186] Embodiments may utilize a high volume of data and may have large data upload and retrieval performance requirements (i.e., performance requirements “prioritizes” performance));
an accuracy mode configured to prioritize accuracy ([0013] The at least one machine learning model relevant to the problem may be further obtained by determining, at the computer system, a combination of the selected and generated models that produces higher accuracy results than the selected and generated models);
an efficiency mode configured to minimize at least one of computational cost and energy ([0175] Embodiments may provide a customer-centric energy system providing improved energy efficiency, cost minimization and reduced CO2 emissions);
a safety mode configured to maximize validation rigor; and
a balanced mode configured to optimize weighted combination of multiple objectives ([0175] Embodiments may provide a customer-centric energy system providing improved energy efficiency, cost minimization and reduced CO2 emissions (i.e., a system that factors in multiple objectives “weighs” those objects at least evenly)).
Regarding claim 5, HOWARD further teaches:
the experiential system is configured to operate without explicit modeling of physical laws, mathematical constraints, or causal mechanisms ([0056] Embodiments of the present systems and methods may provide machine learning techniques that may address such shortcomings and provide improved performance and results. For example, embodiments may address issues in the context of, for example, natural language processing (NLP), in a multidisciplinary approach that aims to bridge the gap between statistical NLP and the many other disciplines necessary for understanding human language such as linguistics, commonsense reasoning, and affective computing. Embodiments may leverage both symbolic and subsymbolic methods as that use models such as semantic networks and conceptual dependency representations to encode meaning, as well as use deep neural networks and multiple kernel learning to infer syntactic patterns from data (i.e., machine learning techniques that generate output using syntactic pattern matching for natural language processing represent operations of an experiential reasoning system, according to [0110] of the specification, that performs commonsense reasoning which does not require explicit modeling of physical laws, mathematical constrains, or causal mechanisms)).
Regarding claim 9, PENFIELD further teaches:
the result validation procedure is operable to compute one or more confidence metrics comprising at least one of: model agreement; constraint satisfaction margins; historical accuracy; uncertainty quantification; and validation results ([Column 18, Line 66-Column 19, Line 9] The output 1235 produced by each pre-trained machine learning model can include individual inferences for each field. Multiple outputs, or inferences may be produced or output 1235 by each network, whereby some are relevant and others may not be. Verification and validation checks can also be applied to either select from candidate outputs or verify and validate the accuracy or relevance of these outputs. After all validation checks are passed, the indexes (or model inferences) from each field are combined and returned and saved in the database 1240 (i.e., passing a validation check represents a “validation result” indicative of a level of “confidence” that the results are valid)).
Regarding claim 10, PENFIELD further teaches:
the result validation procedure is operable to merge a first output received from the [first] system comprising the one or more outputs and a second output received from the [second] system comprising the one or more outputs ([Column 18, Line 66-Column 19, Line 9] The output 1235 produced by each pre-trained machine learning model (i.e., at least a first, experiential model, and second, analytical model as described by HOWARD) can include individual inferences for each field. Multiple outputs, or inferences may be produced or output 1235 by each network, whereby some are relevant and others may not be. Verification and validation checks can also be applied to either select from candidate outputs or verify and validate the accuracy or relevance of these outputs. After all validation checks are passed, the indexes (or model inferences) from each field are combined and returned and saved in the database 1240 (i.e., results from first and second models are validated and combined)).
Regarding claim 12, HOWARD further teaches:
executing a feedback procedure comprising at least one of: adjusting one or more parameters for performing the routing analysis; updating the experiential system; and updating the analytical system ([0081] In embodiments, depending on the complexity of the model and the number of features the algorithm needs to search, the evaluation function can become more elaborate. If there are multiple features for which we want to optimize, a multi-parameter evaluation function can be used, for example a combination of multiple heuristic functions. Then, based on the feedback from all the heuristic functions, a decision can be made concerning how the set of model architectures can be improved).
Regarding claim 13, HOWARD further teaches:
the feedback procedure is performed responsive to at least one of:
outcomes; performance metrics; and user feedback associated with the final result ([0081] In embodiments, depending on the complexity of the model and the number of features the algorithm needs to search, the evaluation function can become more elaborate. If there are multiple features for which we want to optimize, a multi-parameter evaluation function can be used, for example a combination of multiple heuristic functions. Then, based on the feedback from all the heuristic functions, a decision can be made concerning how the set of model architectures can be improved (i.e., heuristic function feedback represents “outcomes” of the heuristic functions)).
Regarding claims 14-18, 22-23, and 25-29, they comprise limitations similar to those of claims 1-5, 9-10, and 12-13, and are therefore rejected for similar rationale.
Regarding claim 30, XU further teaches:
responsive to the result validation procedure determining that the one or more outputs fail to satisfy the one or more physical constraints, the method further comprises performing one or more of: re-routing the user input to a different processing pathway; iteratively refining the one or more outputs until the one or more physical constraints are satisfied ([0158] At block 670, the process iteratively tunes the set of hyperparameters associated with the machine learning model in order to optimize (e.g., obtain an optimal value of the objective function) the machine learning model for the one or more of metrics. [0159] In the process of tuning the hyperparameters, the process evaluates for a current configuration of the hyperparameters (i.e., values of the hyperparameters), a value of the function and determines whether the model is converging and/or a current configuration satisfies each of the one or more constraints… If at least one of the constraints is violated and/or the value of the function is not optimal, the tuning process modifies values of one or more hyperparameters to obtain a new configuration of the hyperparameters and continues training the machine learning model based on the new configuration (i.e., in response to validation determining that a constraint is violated, the process iteratively tunes or “refines” the hyperparameters, thereby refining the output)); and flagging uncertainties and presenting alternative outputs.
Regarding claims 31-32, they comprise limitations similar to claim 30, and are therefore rejected for similar rationale.
Claims 6, and 19 are rejected under 35 U.S.C. 103 as being unpatentable over HOWARD, in view of LEE, in view of ARORA, in view of PENFIELD, in view of XU, as applied to claims 1, and 14 above, and in further view of HYLAND et al. Pub. No.: US 2024/0256796 A1 (hereafter HYLAND)
HYLAND was cited previously.
Regarding claim 6, PENFIELD further teaches:
the analytical system comprises two or more computational implementations of scientific principles directed to: physics; chemistry; biology; economics; and engineering ([Column 10, Line 64-Column 11, Line 2] The custom dataset may be one curated specifically to train a machine learning network to identify specific information. For example, when training a machine learning network such as an NLP model to determine chemical names, CAS numbers, weightings and other information related to chemical ingredients (i.e., machine learning model determines various outputs related to chemistry or chemical engineering principles)).
While HOWARD, in view of LEE, in view of ARORA, in view of PENFIELD, in view of XU discuss an analytical machine learning model implementing scientific principles related at least to chemistry or chemical engineering, HYLAND further teaches:
the analytical system comprises two or more computational implementations of scientific principles directed to: physics; chemistry; biology; economics; and engineering ([0062] training the model using in-domain text comprises performing MLM training; concurrently training the model using labeled general domain task training data, wherein training the model using labeled task training data comprises performing both NLG training and NLU training; and using the trained model to perform a language task within the target domain. [0070] the target domain comprises a domain selected from the list consisting of medical, radiology, biomedical, law, finance, mathematics, chemistry physics, and engineering (i.e., machine learning models are trained, and therefore implement scientific principles directed at least to physics, chemistry, biomedical (biology), finance (economics) and engineering)).
It would have been obvious to a person having ordinary skill in the art before the effective filing date of the invention to have combined HYLAND’s teaching of training machine learning models according to various scientific principles, with HOWARD, in view of LEE, in view of ARORA, in view of PENFIELD, in view of XU’s teaching of operating analytical machine learning models according to scientific principles, to realize, with a reasonable expectation of success, a system that operates analytical machine learning models according to scientific principles, as in HOWARD, in view of LEE, in view of ARORA, in view of PENFIELD, in view of XU, which include physics, chemistry, biomedical, finance, and engineering principles, as in HYLAND. A person having ordinary skill would have been motivated to make this combination to enable machine learning models to operate on a wider array of scientific principles leading to more accurate outputs.
Regarding claim 19, it comprises limitations similar to those of claim 6, and is therefore rejected for similar rationale.
Claims 7, and 20 are rejected under 35 U.S.C. 103 as being unpatentable over HOWARD, in view of LEE, in view of ARORA, in view of PENFIELD, in view of XU, as applied to claims 1, and 14 above, and in further view of LIN et al. Pub. No.: US 2024/0029132 A1 (hereafter LIN)
LIN was cited previously.
Regarding claim 7, while HOWARD, in view of LEE, in view of ARORA, in view of PENFIELD, in view of XU discuss machine learning models that generate output, they do not explicitly teach:
the analytical system is operable to generate an output comprising at least one of:
an uncertainty quantification;
a sensitivity analysis;
a validation certificate;
a documenting constraint satisfaction; and
traceability information.
However, in analogous art that similarly teaches machine learning models generating output, LIN teaches:
the analytical system is operable to generate an output comprising at least one of: an uncertainty quantification; a sensitivity analysis; a validation certificate; a documenting constraint satisfaction; and traceability information ([0045] In some embodiments, the probability output by the machine-learned item availability model 316 includes a confidence score. The confidence score may be an error or uncertainty score of the output availability probability and may be calculated using any standard statistical error measurement).
It would have been obvious to a person having ordinary skill in the art before the effective filing date of the invention to have combined LIN’s teaching of a machine learning model outputting a confidence score indicating an uncertainty quantification, with the combination of HOWARD, in view of LEE, in view of ARORA, in view of PENFIELD, in view of XU’s teaching of an analytical machine learning model generating an output, to realize, with a reasonable expectation of success, a system where an analytical machine learning model generates an output, as in HOWARD, in view of LEE, in view of ARORA, in view of PENFIELD, in view of XU, having an indication of an uncertainty quantification, as in LIN. A person having ordinary skill would have been motivated to make this combination to give an indication of how certain an output is to be accurate for use in making better decisions based on this output.
Regarding claim 20, it comprises limitations similar to those of claim 6, and is therefore rejected for similar rationale.
Claims 8, and 22 are rejected under 35 U.S.C. 103 as being unpatentable over HOWARD, in view of LEE, in view of ARORA, in view of PENFIELD, in view of XU, as applied to claims 1, and 14 above, and in further view of WEINBERGER Patent No.: US 11,227,187 B1 (hereafter WEINBERGER).
WEINBERGER was cited previously.
Regarding claim 7, while HOWARD, in view of LEE, in view of ARORA, in view of PENFIELD, in view of XU discuss a trained machine learning model generating outputs, they do not explicitly teach:
the result validation procedure comprises implementing one or more consistency checking algorithms directed to:
numerical consistency;
logical consistency;
semantic consistency; and
physical consistency.
However in analogous art that similarly teaches a trained machine learning model generating outputs which are validated, WEINBERGER teaches:
the result validation procedure comprises implementing one or more consistency checking algorithms directed to: numerical consistency; logical consistency; semantic consistency; and physical consistency ([Column 20, Lines 7-19] Where the data 475-1, 475-2, 485 is to be used in the generation of a trained machine learning model, the data 475-1, 475-2, 485 may be retrieved and subjected to one or more hash functions or other validation functions. Where the hashes or other values generated based on outputs of the validation functions are consistent with the hashes generated upon authenticating the data 475-1, 475-2, 485, the validity of the data 475-1, 475-2, 485 is confirmed, and a machine learning model may be trained based on the data 475-1, 475-2, 485. If the hashes or other values are not consistent, however, then the validity of the data 475-1, 475-2, 485 is in question, and the data 475-1, 475-2, 485 may not be used to train a machine learning model (i.e., validation directed to hash value consistency represents “numerical consistency”)).
It would have been obvious to a person having ordinary skill in the art before the effective filing date of the invention to have combined WEINBERGER’s teaching of validating machine learning model results based on consistency of numerical hashes, with the combination of HOWARD, in view of LEE, in view of ARORA, in view of PENFIELD, in view of XU’s teaching of generating validated results using machine learning models, to realize, with a reasonable expectation of success, a system that generates validated results using machine learning models, as in HOWARD, in view of LEE, in view of ARORA, in view of PENFIELD, in view of XU, by determining hash value consistency, as in WEINBERGER. A person having ordinary skill would have been motivated to make this combination to better ensure that a machine learning model outputs accurate and desirable results.
Regarding claim 21, it comprises limitations similar to those of claim 6, and is therefore rejected for similar rationale.
Conclusion
THIS ACTION IS MADE FINAL. Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any nonprovisional extension fee (37 CFR 1.17(a)) pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the mailing date of this final action.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to MICHAEL W AYERS whose telephone number is (571)272-6420. The examiner can normally be reached M-F 8:30-5 PM.
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, Aimee Li can be reached at (571) 272-4169. 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.
/MICHAEL W AYERS/Primary Examiner, Art Unit 2195
Prosecution Insights