DETAILED ACTION
This action is responsive to the Application filed on 10/10/2023. Claims 1-20 are pending in the case. Claims 1, 19 and 20 are independent claims.
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 .
Specification
The disclosure is objected to because of the following informalities:
Paragraph [0038], line 7, “…host platform may 102 may…” should read “…host platform 102 may…”
Paragraph [0059], line 20, “…interface 113 and while…” should read “…interface 113 while…”
Paragraph [0060], line 26, “…interface 113 and while…” should read “…interface 113 while…”
Paragraph [0063], lines 13-14, in this example embodiment, the specialized generative AI models have not been introduced yet. Potentially remove “the” from “the specialized generative AI models.”
Appropriate correction is required.
Claim Objections
Claims 17 and 18 are objected to because of the following informalities: The claim sets out to include a list of computer-readable instructions in lines 1-3, but then only recites a condition of the models being trained and no actual step or “instruction” for the processor to carry out. Claim 18 depends on claim 17 and inherits the issue. 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.
The following is a quotation of 35 U.S.C. 112 (pre-AIA ), second paragraph:
The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the applicant regards as his invention.
Claims 1-20 are rejected under 35 U.S.C. 112(b) or 35 U.S.C. 112 (pre-AIA ), second paragraph, as being indefinite for failing to particularly point out and distinctly claim the subject matter which the inventor or a joint inventor (or for applications subject to pre-AIA 35 U.S.C. 112, the applicant), regards as the invention.
Claims 1, 19, and 20 recite the limitation "train, using the normalized portion.” There is insufficient antecedent basis for this limitation in these claims. The examiner notes that there is only sufficient antecedent basis for a single normalized portion of historical information. For continued examination, the examiner comes to interpret “the normalized portions” as “the normalized portion.” Claims 2-18 are dependent on claim 1 and inherit the rejection.
Claim 10 additionally recites, “wherein a plurality of initial generative AI models” in line 1. The reader is unable to discern whether this plurality is “the” plurality of initial generative ai models introduced in claim 7, a new plurality of initial generative ai models, or a portion of “the” plurality of initial generative ai models introduced in claim 7. For continued examination, the examiner comes to interpret “a plurality of initial generative ai models” as “the plurality of initial generative ai models.”
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 1-20 rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
Step 1: Claims 1-18 are directed towards a machine, Claim 19 is directed towards a process, and Claim 20 is directed towards an article of manufacture. Therefore, Claims 1-20 are directed towards one of the 4 statutory categories; process, machine, manufacture or composition of matter.
With respect to claim 1:
Step 2A Prong 1: This claim is directed to a judicial exception.
Select one or more features of the foundational AI model for use in training a plurality of generative AI models (mental process: observations, evaluations, judgements, and opinions. One could perform selection with pen and paper using evaluation)
Identify a portion of the historical information corresponding to the selected one or more features (mental process: observations, evaluations, judgements, and opinions. One could perform identification using observation and evaluation)
Normalize the portion of historical information (Mathematical concept)
Step 2A Prong 2: The judicial exception as a whole is not integrated into a practical application.
Additional Elements:
A computing platform comprising: at least one processor; a communication interface communicatively coupled to the at least one processor; and memory storing computer-readable instructions that, when executed by the at least one processor, cause the computing platform to… (Adding generic computer components to perform the method is not sufficient. 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 using a computer as a tool to perform an abstract idea, as discussed in MPEP § 2106.05(f))
Obtain, from an information storage source, historical information… (Mere data gathering. Insignificant extra-solution activity, as discussed in MPEP § 2106.05(g))
Train, using the historical information, a foundational artificial intelligence (AI) model (High level machine learning. 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 using a computer as a tool to perform an abstract idea, as discussed in MPEP § 2106.05(f))
Train, using the normalized portions of historical information, each generative AI model of the plurality of generative AI models (High level machine learning. 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 using a computer as a tool to perform an abstract idea, as discussed in MPEP § 2106.05(f))
Receive, from a user device, a generative AI prompt (Amounts to necessary data gathering. Insignificant extra-solution activity, as discussed in MPEP § 2106.05(g))
Generate, by inputting the generative AI prompt into one of the plurality of generative AI models, a generative AI response (Generating is a high-level generic function in machine learning. 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 using a computer as a tool to perform an abstract idea, as discussed in MPEP § 2106.05(f))
Send, to the user device, the generative AI response (Amounts to necessary data output. Insignificant extra-solution activity, as discussed in MPEP § 2106.05(g))
Step 2B: The claim does not include additional elements that amount to significantly more than the judicial exception.
Re-evaluation of Insignificant Extra-Solution Activities:
Obtain, from an information storage source, historical information… (“Storing and retrieving information in memory” is a well-understood, routine, conventional activity when claimed in a merely generic manner (as it is in the present claim), as discussed in MPEP § 2106.05(d)(II).
Receive, from a user device, a generative AI prompt (“Receiving or transmitting data over a network” is a well-understood, routine, conventional activity when claimed in a merely generic manner (as it is in the present claim), as discussed in MPEP § 2106.05(d)(II).
Send, to the user device, the generative AI response (“Receiving or transmitting data over a network” is a well-understood, routine, conventional activity when claimed in a merely generic manner (as it is in the present claim), as discussed in MPEP § 2106.05(d)(II).
Additional Elements:
A computing platform comprising: at least one processor; a communication interface communicatively coupled to the at least one processor; and memory storing computer-readable instructions that, when executed by the at least one processor, cause the computing platform to… (Adding generic computer components to perform the method is not sufficient. 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 using a computer as a tool to perform an abstract idea, as discussed in MPEP § 2106.05(f))
Train, using the historical information, a foundational artificial intelligence (AI) model (High level machine learning. 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 using a computer as a tool to perform an abstract idea, as discussed in MPEP § 2106.05(f))
Train, using the normalized portions of historical information, each generative AI model of the plurality of generative AI models (High level machine learning. 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 using a computer as a tool to perform an abstract idea, as discussed in MPEP § 2106.05(f))
Generate, by inputting the generative AI prompt into one of the plurality of generative AI models, a generative AI response (Generating is a high-level generic function in machine learning. 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 using a computer as a tool to perform an abstract idea, as discussed in MPEP § 2106.05(f))
With respect to claim 2:
Step 2A Prong 1: The claim is directed to a judicial exception.
…selecting the one or more features… comprises limiting the foundational AI model to the one or more features (mental process: observations, evaluations, judgements, and opinions. One could perform selection with pen and paper using observation and evaluation)
Step 2A Prong 2: The judicial exception as a whole is not integrated into a practical application.
Additional Elements:
The computing platform of claim 1… (Adding generic computer components to perform the method is not sufficient. 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 using a computer as a tool to perform an abstract idea, as discussed in MPEP § 2106.05(f))
Step 2B: The claim does not include additional elements that amount to significantly more than the judicial exception.
Additional Elements:
The computing platform of claim 1… (Adding generic computer components to perform the method is not sufficient. 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 using a computer as a tool to perform an abstract idea, as discussed in MPEP § 2106.05(f))
With respect to claim 3:
Step 2A Prong 1: The claim is directed to a judicial exception.
…selecting the one or more features comprises selecting structured information features rather than unstructured information features (mental process: observations, evaluations, judgements, and opinions. One could perform selection with pen and paper using observation and evaluation)
Step 2A Prong 2: The judicial exception as a whole is not integrated into a practical application.
Additional Elements:
The computing platform of claim 2… (Adding generic computer components to perform the method is not sufficient. 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 using a computer as a tool to perform an abstract idea, as discussed in MPEP § 2106.05(f))
Step 2B: The claim does not include additional elements that amount to significantly more than the judicial exception.
Additional Elements:
The computing platform of claim 2… (Adding generic computer components to perform the method is not sufficient. 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 using a computer as a tool to perform an abstract idea, as discussed in MPEP § 2106.05(f))
With respect to claim 4:
Step 2A Prong 1: This claim is directed to a judicial exception. Inherits the abstract ideas from claim 2, via dependency.
Step 2A Prong 2: The judicial exception as a whole is not integrated into a practical application.
Additional Elements:
The computing platform of claim 2… (Adding generic computer components to perform the method is not sufficient. 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 using a computer as a tool to perform an abstract idea, as discussed in MPEP § 2106.05(f))
…the portion of the historical information corresponding to the selected one or more features comprises information of the feature limited foundational AI model (Describes the information that the abstract idea operates on rather than an additional element to integrate the exception into a practical application, see MPEP § 2106.05(e), which discusses other meaningful limitations)
Step 2B: The claim does not include additional elements that amount to significantly more than the judicial exception.
Additional Elements:
The computing platform of claim 2… (Adding generic computer components to perform the method is not sufficient. 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 using a computer as a tool to perform an abstract idea, as discussed in MPEP § 2106.05(f))
…the portion of the historical information corresponding to the selected one or more features comprises information of the feature limited foundational AI model (Describes the information that the abstract idea operates on rather than an additional element to integrate the exception into a practical application, see MPEP § 2106.05(e), which discusses other meaningful limitations)
With respect to claim 5:
Step 2A Prong 1: This claim is directed to a judicial exception.
….selecting the one or more features… comprises, for each respective generative AI model, selecting features corresponding to a domain of the corresponding generative AI model (mental process: observations, evaluations, judgements, and opinions. One could perform selection with pen and paper using observation and evaluation)
Step 2A Prong 2: The judicial exception as a whole is not integrated into a practical application.
Additional Elements:
The computing platform of claim 1… (Adding generic computer components to perform the method is not sufficient. 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 using a computer as a tool to perform an abstract idea, as discussed in MPEP § 2106.05(f))
Step 2B: The claim does not include additional elements that amount to significantly more than the judicial exception.
Additional Elements:
The computing platform of claim 1… (Adding generic computer components to perform the method is not sufficient. 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 using a computer as a tool to perform an abstract idea, as discussed in MPEP § 2106.05(f))
With respect to claim 6:
Step 2A Prong 1: This claim is directed to a judicial exception. Inherits the abstract ideas from claim 5, via dependency.
Step 2A Prong 2: The judicial exception as a whole is not integrated into a practical application.
Additional Elements:
The computing platform of claim 5… (Adding generic computer components to perform the method is not sufficient. 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 using a computer as a tool to perform an abstract idea, as discussed in MPEP § 2106.05(f))
Step 2B: The claim does not include additional elements that amount to significantly more than the judicial exception.
Additional Elements:
The computing platform of claim 5… (Adding generic computer components to perform the method is not sufficient. 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 using a computer as a tool to perform an abstract idea, as discussed in MPEP § 2106.05(f))
With respect to claim 7:
Step 2A Prong 1: This claim is directed to a judicial exception.
Limit the selected one or more features…. (Mental process: observations, evaluations, judgements, and opinions. One could perform limitation with pen and paper using evaluation)
…identifying the portion of historical information… comprises identifying a portion of the historical information corresponding to the limited one or more features (mental process: observations, evaluations, judgements, and opinions. One could perform identification with pen and paper using observation)
Step 2A Prong 2: The judicial exception as a whole is not integrated into a practical application.
Additional Elements:
The computing platform of claim 6… (Adding generic computer components to perform the method is not sufficient. 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 using a computer as a tool to perform an abstract idea, as discussed in MPEP § 2106.05(f))
…the memory stores additional computer-readable instructions that, when executed by the at least one processor, further cause the computing platform to… (All generic memories store information including, potentially, instructions. A processor is a generic computer component. 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 using a computer as a tool to perform an abstract idea, as discussed in MPEP § 2106.05(f))
Train, based on the selected one or more features, a plurality of initial generative AI models… (High level machine learning. 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 using a computer as a tool to perform an abstract idea, as discussed in MPEP § 2106.05(f))
Step 2B: The claim does not include additional elements that amount to significantly more than the judicial exception.
Additional Elements:
The computing platform of claim 6… (Adding generic computer components to perform the method is not sufficient. 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 using a computer as a tool to perform an abstract idea, as discussed in MPEP § 2106.05(f))
…the memory stores additional computer-readable instructions that, when executed by the at least one processor, further cause the computing platform to… (All generic memories store information including, potentially, instructions. A processor is a generic computer component. 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 using a computer as a tool to perform an abstract idea, as discussed in MPEP § 2106.05(f))
Train, based on the selected one or more features, a plurality of initial generative AI models… (High level machine learning. 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 using a computer as a tool to perform an abstract idea, as discussed in MPEP § 2106.05(f))
With respect to claim 8:
Step 2A Prong 1: This claim is directed to a judicial exception. Inherits the abstract ideas from claim 7, via dependency.
Step 2A Prong 2: The judicial exception as a whole is not integrated into a practical application.
Additional Elements:
The computing platform of claim 7… (Adding generic computer components to perform the method is not sufficient. 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 using a computer as a tool to perform an abstract idea, as discussed in MPEP § 2106.05(f))
…each of the plurality of initial generative AI models are replaced by a corresponding generative AI model of the plurality of generative AI models (Replacing models with other models does not integrate the exception into a practical application. 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 using a computer as a tool to perform an abstract idea, as discussed in MPEP § 2106.05(f))
Step 2B: The claim does not include additional elements that amount to significantly more than the judicial exception.
Additional Elements:
The computing platform of claim 7… (Adding generic computer components to perform the method is not sufficient. 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 using a computer as a tool to perform an abstract idea, as discussed in MPEP § 2106.05(f))
…each of the plurality of initial generative AI models are replaced by a corresponding generative AI model of the plurality of generative AI models (Replacing models with other models does not integrate the exception into a practical application. 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 using a computer as a tool to perform an abstract idea, as discussed in MPEP § 2106.05(f))
With respect to claim 9:
Step 2A Prong 1: This claim is directed to a judicial exception. Inherits the abstract ideas from claim 7, via dependency.
Step 2A Prong 2: The judicial exception as a whole is not integrated into a practical application.
Additional Elements:
The computing platform of claim 7… (Adding generic computer components to perform the method is not sufficient. 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 using a computer as a tool to perform an abstract idea, as discussed in MPEP § 2106.05(f))
…at least one of the plurality of initial generative AI models is not replaced by a corresponding generative AI model of the plurality of generative AI models (Not replacing one model with another model does not integrate the exception into a practical application. 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 using a computer as a tool to perform an abstract idea, as discussed in MPEP § 2106.05(f))
Step 2B: The claim does not include additional elements that amount to significantly more than the judicial exception.
Additional Elements:
The computing platform of claim 7… (Adding generic computer components to perform the method is not sufficient. 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 using a computer as a tool to perform an abstract idea, as discussed in MPEP § 2106.05(f))
…at least one of the plurality of initial generative AI models is not replaced by a corresponding generative AI model of the plurality of generative AI models (Not replacing one model with another model does not integrate the exception into a practical application. 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 using a computer as a tool to perform an abstract idea, as discussed in MPEP § 2106.05(f))
With respect to claim 10:
Step 2A Prong 1: This claim is directed to a judicial exception. Inherits the abstract ideas from claim 7, via dependency.
Step 2A Prong 2: The judicial exception as a whole is not integrated into a practical application.
Additional Elements:
The computing platform of claim 7… (Adding generic computer components to perform the method is not sufficient. 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 using a computer as a tool to perform an abstract idea, as discussed in MPEP § 2106.05(f))
…a plurality of initial generative AI models include both structured and unstructured information (Describes the information that the abstract idea operates on rather than an additional element to integrate the exception into a practical application, see MPEP § 2106.05(e), which discusses other meaningful limitations)
Step 2B: The claim does not include additional elements that amount to significantly more than the judicial exception.
Additional Elements:
The computing platform of claim 7… (Adding generic computer components to perform the method is not sufficient. 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 using a computer as a tool to perform an abstract idea, as discussed in MPEP § 2106.05(f))
…a plurality of initial generative AI models include both structured and unstructured information (Describes the information that the abstract idea operates on rather than an additional element to integrate the exception into a practical application, see MPEP § 2106.05(e), which discusses other meaningful limitations)
With respect to claim 11:
Step 2A Prong 1: This claim is directed to a judicial exception. Inherits the abstract ideas from claim 1, via dependency.
Step 2A Prong 2: The judicial exception as a whole is not integrated into a practical application.
Additional Elements:
The computing platform of claim 1… (Adding generic computer components to perform the method is not sufficient. 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 using a computer as a tool to perform an abstract idea, as discussed in MPEP § 2106.05(f))
…training the foundational AI model comprises generating a multi-dimensional hyper-space using the historical information (Generating is a high-level generic function in machine learning. 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 using a computer as a tool to perform an abstract idea, as discussed in MPEP § 2106.05(f))
Step 2B: The claim does not include additional elements that amount to significantly more than the judicial exception.
Additional Elements:
The computing platform of claim 1… (Adding generic computer components to perform the method is not sufficient. 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 using a computer as a tool to perform an abstract idea, as discussed in MPEP § 2106.05(f))
…training the foundational AI model comprises generating a multi-dimensional hyper-space using the historical information (Generating is a high-level generic function in machine learning. 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 using a computer as a tool to perform an abstract idea, as discussed in MPEP § 2106.05(f))
With respect to claim 12:
Step 2A Prong 1: This claim is directed to a judicial exception. Inherits the abstract ideas from claim 1, via dependency.
Step 2A Prong 2: The judicial exception as a whole is not integrated into a practical application.
Additional Elements:
The computing platform of claim 11… (Adding generic computer components to perform the method is not sufficient. 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 using a computer as a tool to perform an abstract idea, as discussed in MPEP § 2106.05(f))
…generating the multi-dimensional hyperspace comprises using unsupervised learning to cluster the historical information (Unsupervised learning is a high-level generic function in machine learning. 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 using a computer as a tool to perform an abstract idea, as discussed in MPEP § 2106.05(f))
Step 2B: The claim does not include additional elements that amount to significantly more than the judicial exception.
Additional Elements:
The computing platform of claim 11… (Adding generic computer components to perform the method is not sufficient. 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 using a computer as a tool to perform an abstract idea, as discussed in MPEP § 2106.05(f))
…generating the multi-dimensional hyperspace comprises using unsupervised learning to cluster the historical information (Unsupervised learning is a high-level generic function in machine learning. 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 using a computer as a tool to perform an abstract idea, as discussed in MPEP § 2106.05(f))
With respect to claim 13:
Step 2A Prong 1: This claim is directed to a judicial exception.
…training each generative AI model… comprises clustering the corresponding normalized portion of the historical information… (mental process: observations, evaluations, judgements, and opinions. One could perform clustering on pen and paper using observation and evaluation)
Step 2A Prong 2: The judicial exception as a whole is not integrated into a practical application.
Additional Elements:
The computing platform of claim 11… (Adding generic computer components to perform the method is not sufficient. 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 using a computer as a tool to perform an abstract idea, as discussed in MPEP § 2106.05(f))
…to produce a corresponding heatmap within the multi-dimensional hyper-space (Producing a heat-map within the multi-dimensional hyper-space does not integrate the exception into a practical application. 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 using a computer as a tool to perform an abstract idea, as discussed in MPEP § 2106.05(f))
Step 2B: The claim does not include additional elements that amount to significantly more than the judicial exception.
Additional Elements:
The computing platform of claim 11… (Adding generic computer components to perform the method is not sufficient. 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 using a computer as a tool to perform an abstract idea, as discussed in MPEP § 2106.05(f))
…to produce a corresponding heatmap within the multi-dimensional hyper-space (Producing a heat-map within the multi-dimensional hyper-space does not integrate the exception into a practical application. 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 using a computer as a tool to perform an abstract idea, as discussed in MPEP § 2106.05(f))
With respect to claim 14:
Step 2A Prong 1: This claim is directed to a judicial exception.
…training each generative AI model… comprises: converting the corresponding normalized portion of the historical information to a frequency domain… (mental process: observations, evaluations, judgements, and opinions. One could convert information into a frequency domain with pen and paper using observation and evaluation)
Step 2A Prong 2: The judicial exception as a whole is not integrated into a practical application.
Additional Elements:
The computing platform of claim 1… (Adding generic computer components to perform the method is not sufficient. 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 using a computer as a tool to perform an abstract idea, as discussed in MPEP § 2106.05(f))
Training, using the corresponding converted normalized portion of the historical information, a convolutional neural network (High level machine learning. Recitation of the type of network trained does not integrate the exception into a practical application. 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 using a computer as a tool to perform an abstract idea, as discussed in MPEP § 2106.05(f))
Step 2B: The claim does not include additional elements that amount to significantly more than the judicial exception.
Additional Elements:
The computing platform of claim 1… (Adding generic computer components to perform the method is not sufficient. 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 using a computer as a tool to perform an abstract idea, as discussed in MPEP § 2106.05(f))
Training, using the corresponding converted normalized portion of the historical information, a convolutional neural network (High level machine learning. Recitation of the type of network trained does not integrate the exception into a practical application. 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 using a computer as a tool to perform an abstract idea, as discussed in MPEP § 2106.05(f))
With respect to claim 15:
Step 2A Prong 1: This claim is directed to a judicial exception. Inherits the abstract ideas from claim 14, via dependency.
Step 2A Prong 2: The judicial exception as a whole is not integrated into a practical application.
Additional Elements:
The computing platform of claim 14… (Adding generic computer components to perform the method is not sufficient. 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 using a computer as a tool to perform an abstract idea, as discussed in MPEP § 2106.05(f))
…the convolution neural network is hosted across a plurality of graphics processing units (Hosting the convolution neural network on a plurality of graphics processing units merely indicates the technological environment in which to apply a judicial exception, as discussed in MPEP § 2106.05(h))
Step 2B: The claim does not include additional elements that amount to significantly more than the judicial exception.
Additional Elements:
The computing platform of claim 14… (Adding generic computer components to perform the method is not sufficient. 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 using a computer as a tool to perform an abstract idea, as discussed in MPEP § 2106.05(f))
…the convolution neural network is hosted across a plurality of graphics processing units (Hosting the convolution neural network on a plurality of graphics processing units merely indicates the technological environment in which to apply a judicial exception, as discussed in MPEP § 2106.05(h))
With respect to claim 16:
Step 2A Prong 1: This claim is directed to a judicial exception.
…normalizing the portion of historical information… comprises… subtracting a minimum element value from a value of the given element to produce a first difference, subtracting the minimum element value from a maximum element value to produce a second difference, and dividing the first difference by the second difference (Mathematical concept)
Step 2A Prong 2: The judicial exception as a whole is not integrated into a practical application.
Additional Elements:
The computing platform of claim 1… (Adding generic computer components to perform the method is not sufficient. 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 using a computer as a tool to perform an abstract idea, as discussed in MPEP § 2106.05(f))
Step 2B: The claim does not include additional elements that amount to significantly more than the judicial exception.
Additional Elements:
The computing platform of claim 1… (Adding generic computer components to perform the method is not sufficient. 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 using a computer as a tool to perform an abstract idea, as discussed in MPEP § 2106.05(f))
With respect to claim 17:
Step 2A Prong 1: This claim is directed to a judicial exception. Inherits the abstract ideas from claim 7, via dependency.
Step 2A Prong 2: The judicial exception as a whole is not integrated into a practical application.
Additional Elements:
The computing platform of claim 10… (Adding generic computer components to perform the method is not sufficient. 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 using a computer as a tool to perform an abstract idea, as discussed in MPEP § 2106.05(f))
…the memory stores additional computer-readable instructions that, when executed by the at least one processor, further cause the computing platform to… (All generic memories store information including, potentially, instructions. A processor is a generic computer component. 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 using a computer as a tool to perform an abstract idea, as discussed in MPEP § 2106.05(f))
…the plurality of generative AI models includes at least a first generative AI model and a second generative AI model, wherein the first generative AI model is directed to a first domain and the second generative AI model is directed to a second domain, different that the first domain (Describes the information that the abstract idea operates on rather than an additional element to integrate the exception into a practical application, see MPEP § 2106.05(e), which discusses other meaningful limitations)
Step 2B: The claim does not include additional elements that amount to significantly more than the judicial exception.
Additional Elements:
The computing platform of claim 10… (Adding generic computer components to perform the method is not sufficient. 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 using a computer as a tool to perform an abstract idea, as discussed in MPEP § 2106.05(f))
…the memory stores additional computer-readable instructions that, when executed by the at least one processor, further cause the computing platform to… (All generic memories store information including, potentially, instructions. A processor is a generic computer component. 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 using a computer as a tool to perform an abstract idea, as discussed in MPEP § 2106.05(f))
…the plurality of generative AI models includes at least a first generative AI model and a second generative AI model, wherein the first generative AI model is directed to a first domain and the second generative AI model is directed to a second domain, different that the first domain (Describes the information that the abstract idea operates on rather than an additional element to integrate the exception into a practical application, see MPEP § 2106.05(e), which discusses other meaningful limitations)
With respect to claim 18:
Step 2A Prong 1: This claim is directed to a judicial exception. Inherits the abstract ideas from claim 7, via dependency.
Step 2A Prong 2: The judicial exception as a whole is not integrated into a practical application.
Additional Elements:
The computing platform of claim 17… (Adding generic computer components to perform the method is not sufficient. 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 using a computer as a tool to perform an abstract idea, as discussed in MPEP § 2106.05(f))
…a unique heatmap is produced for each of the first generative AI model and the second generative AI model within a multi-dimensional hyperspace (Producing a heat-map for each model within the multi-dimensional hyper-space does not integrate the exception into a practical application. 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 using a computer as a tool to perform an abstract idea, as discussed in MPEP § 2106.05(f))
Step 2B: The claim does not include additional elements that amount to significantly more than the judicial exception.
Additional Elements:
The computing platform of claim 17… (Adding generic computer components to perform the method is not sufficient. 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 using a computer as a tool to perform an abstract idea, as discussed in MPEP § 2106.05(f))
…a unique heatmap is produced for each of the first generative AI model and the second generative AI model within a multi-dimensional hyperspace (Producing a heat-map for each model within the multi-dimensional hyper-space does not integrate the exception into a practical application. 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 using a computer as a tool to perform an abstract idea, as discussed in MPEP § 2106.05(f))
With respect to claim 19:
Step 2A Prong 1: This claim is directed to a judicial exception.
Selecting one or more features of the foundational AI model for use in training a plurality of generative AI models (mental process: observations, evaluations, judgements, and opinions. One could perform selection with pen and paper using evaluation)
Identifying a portion of the historical information corresponding to the selected one or more features (mental process: observations, evaluations, judgements, and opinions. One could perform identification using observation and evaluation)
Normalizing the portion of historical information (Mathematical concept)
Step 2A Prong 2: The judicial exception as a whole is not integrated into a practical application.
Additional Elements:
…at a computing platform comprising at least one processor, a communication interface, and memory… (Adding generic computer components to perform the method is not sufficient. 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 using a computer as a tool to perform an abstract idea, as discussed in MPEP § 2106.05(f))
Obtaining, from an information storage source, historical information… (Mere data gathering. Insignificant extra-solution activity, as discussed in MPEP § 2106.05(g))
Training, using the historical information, a foundational artificial intelligence (AI) model (High level machine learning. 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 using a computer as a tool to perform an abstract idea, as discussed in MPEP § 2106.05(f))
Training, using the normalized portions of historical information, each generative AI model of the plurality of generative AI models (High level machine learning. 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 using a computer as a tool to perform an abstract idea, as discussed in MPEP § 2106.05(f))
Receiving, from a user device, a generative AI prompt (Amounts to necessary data gathering. Insignificant extra-solution activity, as discussed in MPEP § 2106.05(g))
Generating, by inputting the generative AI prompt into one of the plurality of generative AI models, a generative AI response (Generating is a high-level generic function in machine learning. 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 using a computer as a tool to perform an abstract idea, as discussed in MPEP § 2106.05(f))
Sending, to the user device, the generative AI response (Amounts to necessary data output. Insignificant extra-solution activity, as discussed in MPEP § 2106.05(g))
Step 2B: The claim does not include additional elements that amount to significantly more than the judicial exception.
Re-evaluation of Insignificant Extra-Solution Activities:
Obtaining, from an information storage source, historical information… (“Storing and retrieving information in memory” is a well-understood, routine, conventional activity when claimed in a merely generic manner (as it is in the present claim), as discussed in MPEP § 2106.05(d)(II).
Receiving, from a user device, a generative AI prompt (“Receiving or transmitting data over a network” is a well-understood, routine, conventional activity when claimed in a merely generic manner (as it is in the present claim), as discussed in MPEP § 2106.05(d)(II).
Sending, to the user device, the generative AI response (“Receiving or transmitting data over a network” is a well-understood, routine, conventional activity when claimed in a merely generic manner (as it is in the present claim), as discussed in MPEP § 2106.05(d)(II).
Additional Elements:
…at a computing platform comprising at least one processor, a communication interface, and memory… (Adding generic computer components to perform the method is not sufficient. 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 using a computer as a tool to perform an abstract idea, as discussed in MPEP § 2106.05(f))
Training, using the historical information, a foundational artificial intelligence (AI) model (High level machine learning. 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 using a computer as a tool to perform an abstract idea, as discussed in MPEP § 2106.05(f))
Training, using the normalized portions of historical information, each generative AI model of the plurality of generative AI models (High level machine learning. 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 using a computer as a tool to perform an abstract idea, as discussed in MPEP § 2106.05(f))
Generating, by inputting the generative AI prompt into one of the plurality of generative AI models, a generative AI response (Generating is a high-level generic function in machine learning. 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 using a computer as a tool to perform an abstract idea, as discussed in MPEP § 2106.05(f))
With respect to claim 20:
Step 2A Prong 1: This claim is directed to a judicial exception.
Selecting one or more features of the foundational AI model for use in training a plurality of generative AI models (mental process: observations, evaluations, judgements, and opinions. One could perform selection with pen and paper using evaluation)
Identifying a portion of the historical information corresponding to the selected one or more features (mental process: observations, evaluations, judgements, and opinions. One could perform identification using observation and evaluation)
Normalizing the portion of historical information (Mathematical concept)
Step 2A Prong 2: The judicial exception as a whole is not integrated into a practical application.
Additional Elements:
One or more non-transitory computer-readable media storing instructions… (non-transitory computer-readable media are generic computer components that all store information including, potentially, instructions. 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 using a computer as a tool to perform an abstract idea, as discussed in MPEP § 2106.05(f))
…executed by a computing platform comprising at least one processor, a communication interface, and memory, cause the computing platform to… (Adding generic computer components to perform the method is not sufficient. 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 using a computer as a tool to perform an abstract idea, as discussed in MPEP § 2106.05(f))
Obtaining, from an information storage source, historical information… (Mere data gathering. Insignificant extra-solution activity, as discussed in MPEP § 2106.05(g))
Training, using the historical information, a foundational artificial intelligence (AI) model (High level machine learning. 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 using a computer as a tool to perform an abstract idea, as discussed in MPEP § 2106.05(f))
Training, using the normalized portions of historical information, each generative AI model of the plurality of generative AI models (High level machine learning. 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 using a computer as a tool to perform an abstract idea, as discussed in MPEP § 2106.05(f))
Receiving, from a user device, a generative AI prompt (Amounts to necessary data gathering. Insignificant extra-solution activity, as discussed in MPEP § 2106.05(g))
Generating, by inputting the generative AI prompt into one of the plurality of generative AI models, a generative AI response (Generating is a high-level generic function in machine learning. 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 using a computer as a tool to perform an abstract idea, as discussed in MPEP § 2106.05(f))
Sending, to the user device, the generative AI response (Amounts to necessary data output. Insignificant extra-solution activity, as discussed in MPEP § 2106.05(g))
Step 2B: The claim does not include additional elements that amount to significantly more than the judicial exception.
Re-evaluation of Insignificant Extra-Solution Activities:
Obtaining, from an information storage source, historical information… (“Storing and retrieving information in memory” is a well-understood, routine, conventional activity when claimed in a merely generic manner (as it is in the present claim), as discussed in MPEP § 2106.05(d)(II).
Receiving, from a user device, a generative AI prompt (“Receiving or transmitting data over a network” is a well-understood, routine, conventional activity when claimed in a merely generic manner (as it is in the present claim), as discussed in MPEP § 2106.05(d)(II).
Sending, to the user device, the generative AI response (“Receiving or transmitting data over a network” is a well-understood, routine, conventional activity when claimed in a merely generic manner (as it is in the present claim), as discussed in MPEP § 2106.05(d)(II).
Additional Elements:
One or more non-transitory computer-readable media storing instructions… (non-transitory computer-readable media are generic computer components that all store information including, potentially, instructions. 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 using a computer as a tool to perform an abstract idea, as discussed in MPEP § 2106.05(f))
…executed by a computing platform comprising at least one processor, a communication interface, and memory, cause the computing platform to… (Adding generic computer components to perform the method is not sufficient. 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 using a computer as a tool to perform an abstract idea, as discussed in MPEP § 2106.05(f))
Training, using the historical information, a foundational artificial intelligence (AI) model (High level machine learning. 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 using a computer as a tool to perform an abstract idea, as discussed in MPEP § 2106.05(f))
Training, using the normalized portions of historical information, each generative AI model of the plurality of generative AI models (High level machine learning. 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 using a computer as a tool to perform an abstract idea, as discussed in MPEP § 2106.05(f))
Generating, by inputting the generative AI prompt into one of the plurality of generative AI models, a generative AI response (Generating is a high-level generic function in machine learning. 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 using a computer as a tool to perform an abstract idea, as discussed in MPEP § 2106.05(f))
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.
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 1-7, 10, 17, and 19-20 are rejected under 35 U.S.C 103 as being unpatentable over Rice et al. (US 20220269927 A1), in view of Bhamidipaty et al. (US 20230029218 A1).
Regarding claim 1, Rice teaches a computing platform (Fig. 20, Ref No. 4700) comprising:
at least one processor (Fig. 20, Ref No. 4702);
a communication interface communicatively coupled to the at least one processor (Fig. 20 and Paragraph [0158], “In particular embodiments, bus 4712 includes hardware, software, or both coupling components of computer system 4700 to each other”); and
memory (Fig. 20, Ref No. 4704) storing computer-readable instruction that, when executed by the at least one processor, cause the computing platform to (Paragraph [0153], “In particular embodiments, processor 4702 executes only instructions in one or more internal registers or internal caches or in memory 4704“):
obtain, from an information storage source, historical information… (Paragraph [0153], “and operates only on data in one or more internal registers or internal caches or in memory 4704” and Fig. 4, Ref No. 1420)
train, using the historical information, a foundational artificial intelligence (AI) model (Fig. 4, Ref No. 1420, which depicts an embodiment that trains a base model for which will later be partitioned. Note that the only stipulations made in the specification regarding the form of the foundational AI model were that it should be generalized and hold a massive amount of information covering multiple things. Paragraph [0083], “In particular embodiments, the system may allow feature engineering to effectively handle a huge amount of data for a specific problem domain or for multiple problem domains”… “allow the feature engineering to be performed in a large-scale domain” … “the system may create better features for better models and feed the optimized features to ML models”);
select one or more features of the foundational AI model for use in training a plurality of generative AI models (Fig. 5 and Paragraph [0058], describes the process of using feature engineering to determine features for a ML model and then using those features in the modeling and training process to create or train the ML model based on the features, allowing for the corresponding ML models (plurality) to be evaluated. Note that with no special definition generative AI models, under their broadest reasonable interpretation, are any AI models capable of generating any form of entirely new outputs. Paragraph [0119], “The system may generate recommendations for useful models to users who own some newly uploaded data,” other examples can be seen throughout the disclosure);
identify a portion of the historical information corresponding to the selected one or more features (Paragraph [0060], “Data and features may be divided into different layers, such as, fundamental data (through pre-processing raw data), generic features, domain features, and application features”, which involves identifying portions of data that correspond to different features);
train, using the normalized portions of the historical information, each generative AI model of the plurality of generative AI models (Fig. 9, depicts determining features based on different data layers and training one or more ML models based on those features);
…receive, from a user device, a generative AI prompt; generate, by inputting the generative AI prompt into one of the plurality of generative AI models, a generative AI response; and send, to the user device, the generative AI response (Paragraph [0068], describes an embodiment in which the systems inference services might use a user input/output module to allow the user to query the system about a particular ML model (in the plurality of models within the system) and then generate recommended features for the ML model based on a knowledge graph or relationship)
Rice et al. does not distinctly disclose that the historical information obtained is comprised of both structured and unstructured information or that the historical information is normalized prior to training.
However, Bhamidipaty et al. teaches training using historical information, wherein the historical information includes both structured and unstructured information (Paragraph [0055], “At 412, the built machine learning model using both the structured and unstructured data can be run”) and
normalizing the portion of the historical information used in training (Paragraph [0029], “At 114, a processor may perform variable formulation and selection, e.g., using domain expertise, normalization, and correlation analyses,” using these formulated variables in machine learning modeling)
Before the effective filing date of the claimed invention, it would have been obvious to one of ordinary skill in the art to combine the system of Rice with the techniques of using heterogenous data sources and normalizing the data in Bhamidipaty to augment the feature set and standardize feature scales to common range in order to improve the accuracy of the models (Bhamidipaty, Paragraph [0020], lines 1-3).
Regarding claim 2, Rice as modified by Bhamidipaty teaches all of the limitations of the computing platform in claim 1 as cited above and Rice further teaches the part of the feature selection process:
wherein selecting one or more features of the foundational AI model for use in training a plurality of generative AI models comprises limiting the foundational AI model to the one or more features (Paragraph [0058], “The determined features may be used in a modeling and training process 2103 to create a ML model or/and train the ML model based on these features,” thus limiting the base model to these selected features)
Regarding claim 3, Rice as modified by Bhamidipaty teaches all of the limitations of the computing platform in claim 2 as cited above and Bhamidipaty further teaches the part of the feature selection process:
wherein selecting the one or more features comprises selecting structured information features rather than unstructured information features (Paragraph [0055], “A machine learning model using only the structured data ( e.g., from 404) can also be built and run.” Fig. 4 also depicts the potential to create a structured feature only model at Ref No. 402 or Ref No. 410 if 406 and 408 are bypassed)
Although the rationale in the rejection of claim 1 as seen above does not cover this limitation, it can be further seen that before the effective filing date of the claimed invention it would have been obvious to one of ordinary skill in the art to combine the system of Rice with the techniques for selecting structured information rather than unstructured information in Bhamidipaty to allow for the creation of models geared towards specific tasks (Bhamidipaty, Paragraph [0053]) and allow models trained on different types of data to be compared on their performance (Bhamidipaty, Paragraph [0055]).
Regarding claim 4, Rice as modified by Bhamidipaty teaches all of the limitations of the computing platform in claim 2 as cited above and Rice further teaches the part of the information selection process:
wherein the portion of the historical information corresponding to the selected one or more features comprises information of the feature limited foundational AI model (Paragraph [0060], “particular embodiments of the system may allow the known features of the mass feature repository to be reused in different ML models and allow knowledge of the existing models to be used to new knowledge domain.” This is also a given because the information corresponding to the selected features is the same as the information in the limited foundational model when the selected features and limited features are the same, as is the case)
Regarding claim 5, Rice as modified by Bhamidipaty teaches all of the limitations of the computing platform in claim 1 as cited above and Rice further teaches the part of the feature selection process:
wherein selecting the one or more features of the foundational AI model for use in training a plurality of generative AI models comprises, for each respective generative AI model selecting features corresponding to a domain of the corresponding generative AI model (Paragraph [0058], “As an example, feature engineering may be performed by human experts to determine features for a ML model based on the domain knowledge associated with that ML model”)
Regarding claim 6, Rice as modified by Bhamidipaty teaches all of the limitations of the computing platform in claim 5 as cited above and Bhamidipaty further teaches the part of the feature selection process:
wherein, the selected one or more features of the foundational AI model include both structured information features and unstructured information features (Paragraph [0055], “At 410, one or more of the structured data (e.g., from 404) and one or more of the unstructured data (e.g., from 408) can be used as features for building a machine learning model. At 412, the built machine learning model using both the structured and unstructured data can be run,
which generates an output)
See the rationale for combining Rice and Bhamidipaty in the rejection for claim 1 above.
Regarding claim 7, Rice as modified by Bhamidipaty teaches all of the limitations of the computing platform in claim 6 as cited above and Rice further teaches the instructions in memory:
Wherein, the memory stores additional computer-readable instructions that when executed by the at least one processor, further cause the computing platform to: train, based on the selected one or more features, a plurality of initial generative AI models (Paragraph [0066], “One or more processors 602 may execute computer instructions stored in memory
604 and Fig. 9, depicts determining features based on different data layers and training one or more ML models based on those features)
While Bhamidipaty further teaches the limitation of these models:
…and limit the selected one or more features of the plurality of initial generative AI models to include only structured information features, wherein identifying the portion of the historical information corresponding to the selected one or more features comprises identifying a portion of the historical information corresponding to the limited one or more features (Fig. 4, depicts the process of training a combined model created for structured and unstructured information wherein the combined trained model could potentially only select the structured information features for the final model. In Ref No. 402 it can be seen that structured data sources are selected which correspond to the limited structured features)
See the rationale for combining Rice and Bhamidipaty in the rejection for claim 3 above.
Regarding claim 10, Rice as modified by Bhamidipaty teaches all of the limitations of the computing platform in claim 7 as cited above and Bhamidipaty further teaches the part of the training process:
Wherein, a plurality of initial generative AI models include both structured and unstructured information (Fig. 4)
See the rationale for combining Rice and Bhamidipaty in the rejection for claim 1 above.
Regarding claim 17, Rice as modified by Bhamidipaty teaches all of the limitations of the computing platform in claim 10 as cited above and Rice further teaches the part of the system:
wherein the memory stores additional computer-readable instructions that, when executed by the at least one processor, further cause the computing platform to: wherein the plurality of generative AI models includes at least a first generative AI model and a second generative AI model, wherein the first generative AI model is directed to a first domain and the second generative AI model is directed to a second domain, different than the first domain. (Paragraph [0080], “In particular embodiments, the system may use feature engineering technology to generate features that can be used to train machine-learning (ML) models cross different problem domains”).
Regarding claim 19, see the rejection for claim 1 above. Note that the only difference between claim 19 and claim 1 is that claim 19 is directed to the method whereas claim 1 is directed to the system that carries out the method, i.e. covering the method.
Regarding claim 20, see the rejection for claim 1 above. Note that the only difference between claim 20 and claim 1 is that claim 20 is directed to the article of manufacture whereas claim 1 is directed to the system that uses that article of manufacture, i.e. covering the article of manufacture.
Claims 8 and 9 are rejected under 35 U.S.C 103 as being unpatentable over Rice et al. (US 20220269927 A1), in view of Bhamidipaty et al. (US 20230029218 A1), further in view of Mai et al. (Mai, Gengchen, et al. "Towards a foundation model for geospatial artificial intelligence (vision paper)." Proceedings of the 30th International Conference on Advances in Geographic Information Systems. 2022.).
Regarding claim 8, Rice as modified by Bhamidipaty teaches all of the limitations of the computing platform in claim 7 as cited above, but does not distinctly disclose part of the training process:
wherein each of the plurality of initial generative AI models are replaced by a corresponding generative AI model of the plurality of generative AI models.
However, Mai et al. teaches that portion of the training process:
wherein each of the plurality of initial generative AI models are replaced by a corresponding generative AI model of the plurality of generative AI models. (Abstract, “We first show the advantages of this idea by testing the performance of existing Large pre-trained Language Models (LLMs) (e.g. GPT-2 and GPT-3) on two geospatial semantics tasks. Results indicate that these task-agnostic LLMs can outperform task-specific fully-supervised models on both tasks with 2-9% improvement in a few-shot learning setting”)
Before the effective filing date of the claimed invention it would have been obvious to one of ordinary skill in the art to modify the system as taught by Rice modified by the techniques taught by Bhamidipaty to include the replacement of limited models with unlimited models as taught by Mai in order to improve performance on specific tasks in few-shot learning environments (Mai, Abstract) and allow the models to be deployed across a wider range of domains, sharing knowledge among said domains, and limiting the need for task-specific training data (Main, Introduction).
Regarding claim 9, Rice modified by Bhamidipaty teaches all of the limitations of the computing platform in claim 7 as cited above, but does not distinctly disclose part of the training process:
Wherein at least one of the plurality of initial generative AI models is not replaced by a corresponding generative AI model of the plurality of generative AI models
However, Mai et al. teaches that portion of the training process:
Wherein at least one of the plurality of initial generative AI models is not replaced by a corresponding generative AI model of the plurality of generative AI models (Pre-trained Language Models Hold Promise for GeoAI, the experiments used provide basis for allowing the specific tasks to be performed by the foundational models and the task-specific models, thereby not replacing every task-specific model with a corresponding foundational LLM.)
Although the rationale for combining Rice as modified by Bhamidipaty with Mai, as seen in the rejection for claim 8 above, does not cover this limitation, it can be further seen that before the effective filing date of the claimed invention it would have been obvious to one of ordinary skill in the art to modify the system as taught by Rice modified by the techniques taught by Bhamidipaty to include the non-replacement of at least one model as taught by Mai in order to prevent amplification of biases in data rich regions when using non-limited models, make the models easier to interpret, and prevent down-stream models from inheriting biases (Risks and Challenges, Paragraph 1).
Claims 11 and 12 are rejected under 35 U.S.C 103 as being unpatentable over Rice et al. (US 20220269927 A1), in view of Bhamidipaty et al. (US 20230029218 A1), further in view of Xia et al. (CN 113850393 A).
Regarding claim 11, Rice as modified by Bhamidipaty teaches all of the limitations of the computing platform in claim 1 as cited above, but does not distinctly disclose part of the training process:
Wherein training the foundational AI model comprises generating a multi-dimensional hyper-space using the historical information
However, Xia et al. teaches that portion of the training process:
Wherein training the foundational AI model comprises generating a multi-dimensional hyper-space using the historical information (Detailed Description of the Embodiments, Paragraph 8, “In a specific embodiment, the knowledge graph described in the present disclosure may be a data structure comprising knowledge and information to represent (e.g., represented by the node) interaction between each entity (e.g., interaction represented by the side)”. A multi-dimensional hyper-space falls under this definition of a knowledge graph)
Before the effective filing date of the claimed invention it would have been obvious to one of ordinary skill in the art to modify the system as taught by Rice modified by the techniques taught by Bhamidipaty to include the generation of a multi-dimensional hyperspace of the historical information during training as taught by Xia in order to provide the models with a way to learn new knowledge and characteristics of the data (Xia, Overview of Specific Embodiment, Paragraph 1) and then use that new knowledge to recommend new features that would improve the accuracy and precision of the models (Xia, Overview of Specific Embodiment, Paragraph 2).
Regarding claim 12, Rice as modified by Bhamidipaty, further modified by Xia, teaches all of the limitations of the computing platform in claim 11 as cited above and Xia further teaches the part of training:
Wherein, generating the multi-dimensional hyper-space comprises using unsupervised learning to cluster the historical information (Description of Example Embodiments, Paragraph 17, “ In particular embodiments, the system may access features in the knowledge graph (e.g., using random walk or deep walk), cluster the features into a plurality of feature categories or feature groups, and merge similar features or similar feature groups to reduce the number of features and simplify the knowledge graph,” both random walk and deep walk are examples of unsupervised learning as is known to anyone ordinarily skilled in the art).
See the rationale for combining Rice as modified by Bhamidipaty with Xia in the rejection for claim 11 above.
Claim 13 is rejected under 35 U.S.C 103 as being unpatentable over Rice et al. (US 20220269927 A1), in view of Bhamidipaty et al. (US 20230029218 A1), further in view of Xia et al. (CN 113850393 A), further in view of Senbabaoglu et al. (Senbabaoglu, Yasin, George Michailidis, and Jun Z. Li. "A reassessment of consensus clustering for class discovery “. bioRxiv (2014): 002642.)
Regarding claim 13, Rice as modified by Bhamidipaty, further modified by Xia, teaches all of the limitations of the computing platform in claim 11 as cited above, but does not distinctly disclose the part of the training process:
Wherein training each generative AI model of the plurality of generative AI models comprises clustering the corresponding normalized portion of the historical information to produce a corresponding heatmap within the multi-dimensional hyper-space
However, Senbabaoglu et al. teaches that portion of the training process:
Wherein training each generative AI model of the plurality of generative AI models comprises clustering the corresponding normalized portion of the historical information to produce a corresponding heatmap within the multi-dimensional hyper-space (Gene-gene Correlation Among Most Discriminant Genes Makes It Easy to “Validate” Any K, Paragraph 1, “In a popular implementation [2,17], the best classifier genes for each of K clusters are chosen from the learning set, and a heatmap of all learning samples with only these genes is constructed, with the samples and the genes both grouped in K clusters, “ and Quantitative comparisons of Cluster Strength Between GBM1 and the Null Datasets, Paragraph 4, “This underlines the fact that different clustering measures emphasize different features of a given heterogeneous high-dimensional dataset).
Before the effective filing date of the claimed invention it would have been obvious to one of ordinary skill in the art to modify the system as taught by Rice modified by the techniques taught in Bhamidipaty and Xia to include the clustering of information and creation of heatmaps as taught by Senbabaoglu in order to allow the models to visualize the clusters and their cluster strengths so as to improve each models accuracy through more effective training (Senbabaoglu, Quantitative Comparisons of Cluster Strength Between GBM1 and the Null Datasets, “CC heatmaps in Figure 1 and 5 allow visual comparisons of cluster strength”).
Claim 14 and 15 are rejected under 35 U.S.C 103 as being unpatentable over Rice et al. (US 20220269927 A1), in view of Bhamidipaty et al. (US 20230029218 A1), further in view of Cheng et al. (CN 115017957 A).
Regarding claim 14, Rice as modified by Bhamidipaty teaches all of the limitations of the computing platform in claim 1 as cited above, but does not distinctly disclose the part of the training process:
wherein training each generative AI model of the plurality of generative AI models comprises: converting the corresponding normalized portion of the historical information to a frequency domain; and
training, using the corresponding converted normalized portion of the historical information, a convolutional neural network.
However, Cheng et al. teaches that part of the training process:
wherein training each generative AI model of the plurality of generative AI models comprises: converting the corresponding normalized portion of the historical information to a frequency domain; and training, using the corresponding converted normalized portion of the historical information, a convolutional neural network (Detailed Description, Paragraph 24, “By converting the AE signals into a speech spectrogram of a frequency domain nature, the attributes of the phonemes may be better identified by the speech spectrogram. The obtained signal is used as a characteristic and input into a CNN (Convolutional Neural Networks) for model training, and the AE signal identification accuracy rate is improved)
Before the effective filing date of the claimed invention, it would have been obvious to one of ordinary skill in the art to modify the system as taught by Rice modified by the techniques taught by Bhamidipaty to include the conversion of the data into a frequency domain and use of said frequency domain to train a CNN as taught by Cheng in order to allow certain attributes of the data to be discerned via the frequency domain (Cheng, Detailed Description, Paragraph 24, “the attributes of the phonemes may be better identified by the speech spectrogram”) and improve the accuracy of signal identification (Cheng, Detailed Description, Paragraph 24, “AE signal identification accuracy rate is improved”), thereby improving the accuracy of the CNN.
Regarding claim 15, Rice as modified by Bhamidipaty, further modified by Cheng, teaches all of the limitations of the computing platform in claim 14 as cited above and Rice further teaches the design of the system:
Wherein, the convolutional neural network is hosted across a plurality of graphics processing units (Paragraph [0026], “Particular embodiments of the system may break single host (GPU/CPU) memory limit by using pipeline parallelism instead of simple model parallelism”)
Claim 16 is rejected under 35 U.S.C 103 as being unpatentable over Rice et al. (US 20220269927 A1), in view of Bhamidipaty et al. (US 20230029218 A1), further in view of Cabello-Solorzano et al. (Cabello-Solorzano, Kelsy, et al. "The impact of data normalization on the accuracy of machine learning algorithms: A comparative analysis." International conference on soft computing models in industrial and environmental applications. Cham: Springer Nature Switzerland, 2023.).
Regarding claim 16, Rice as modified by Bhamidipaty teaches all of the limitations of the computing platform in claim 1 as cited above, but does not distinctly disclose the part of the training process:
Wherein normalizing the portion of the historical information comprises, for each element of the portion of the historical information: subtracting a minimum element value from a value of the given element to produce a first difference, subtracting the minimum element value from a maximum element value to produce a second difference, and dividing the first difference by the second difference.
However, Cabello-Solorzano et al. teaches that part of the training process:
Wherein normalizing the portion of the historical information comprises, for each element of the portion of the historical information: subtracting a minimum element value from a value of the given element to produce a first difference, subtracting the minimum element value from a maximum element value to produce a second difference, and dividing the first difference by the second difference. (Background, Min-Max Normalization, Eq. 1)
Before the effective filing date of the claimed invention, it would have been obvious to one of ordinary skill in the art to modify the system as taught by Rice modified by the techniques taught by Bhamidipaty to include the min-max method of normalization as taught Cabello-Solorzano in order to scale the data to a common range so that analysis and comparison of variables becomes easier for the models (Cabello-Solorzano, Background, Min-Max Normalization, Paragraph 2).
Claim 18 is rejected under 35 U.S.C 103 as being unpatentable over Rice et al. (US 20220269927 A1), in view of Bhamidipaty et al. (US 20230029218 A1), further in view of Senbabaoglu et al. (Senbabaoglu, Yasin, George Michailidis, and Jun Z. Li. "A reassessment of consensus clustering for class discovery “. bioRxiv (2014): 002642.)
Regarding claim 18, Rice as modified by Bhamidipaty teaches all of the limitations of the computing platform in claim 17 as cited above, but does not distinctly disclose the portion of training:
wherein a unique heatmap is produced for each of the first generative AI model and the second generative AI model within a multidimensional hyper-space
However, Senbabaoglu et al. teaches this part of the training process:
wherein a unique heatmap is produced for each of the first generative AI model and the second generative AI model within a multidimensional hyper-space (Gene-gene Correlation Among Most Discriminant Genes Makes It Easy to “Validate” Any K, Paragraph 1, “In a popular implementation [2,17], the best classifier genes for each of K clusters are chosen from the learning set, and a heatmap of all learning samples with only these genes is constructed, with the samples and the genes both grouped in K clusters, “ and Quantitative comparisons of Cluster Strength Between GBM1 and the Null Datasets, Paragraph 4, “This underlines the fact that different clustering measures emphasize different features of a given heterogeneous high-dimensional dataset)
See the rationale for modifying Rice, as modified by Bhamidipaty and Xia, to include the teachings of Senbabaoglu in the rejection for claim 13 above.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to Zane A Rawlings whose telephone number is (571) 270-3372. The examiner can normally be reached M-F, 8am to 5pm ET.
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, Alexey Shmatov can be reached at (571) 270-3428. 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.
/Z.A.R./Examiner, Art Unit 2123
/ALEXEY SHMATOV/Supervisory Patent Examiner, Art Unit 2123
Prosecution Insights