DETAILED ACTION
Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
This action is made final.
Claims 1, 2, 4, 5, 7 and 8 are pending. Claims 1, 4 and 7 are independent claims.
Response to Arguments
Applicant’s arguments, dated 1/22/2026, regarding the 35 U.S.C. 101 rejections of the previous office action have been fully considered but are unpersuasive. The scope of the claims has changed – see the updated rejection below. In response to applicant's argument that the references fail to show certain features of the invention, it is noted that the features upon which applicant relies (i.e., AUC maximization) are not recited in the rejected claim(s). Although the claims are interpreted in light of the specification, limitations from the specification are not read into the claims. See In re Van Geuns, 988 F.2d 1181, 26 USPQ2d 1057 (Fed. Cir. 1993). Applicant argues that filtering data and the pair generation process are not mathematical calculations, however, the examiner described these processes as being mental. Applicant argues that the claims generally integrate any possible judicial exception into a practical application, without pointing out specific claim limitations that provide a benefit or additional elements that, when considered alone or in combination, integrate a potential judicial exception into a practical application. Examiner argues that the technical benefit cannot be provided by the judicial exception (i.e., abstract ideas), only by additional elements and additional elements in combination with the judicial exception.
Applicant’s arguments, dated 1/22/2026, regarding the 35 U.S.C. 103 rejections of the previous office action have been fully considered but are unpersuasive. The scope of the claims has changed – see the updated rejection below.
Claim Rejections - 35 USC § 101
35 U.S.C. 101 reads as follows:
Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.
Claims 1, 2, 4, 5, 7 and 8 are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
Regarding claim 1:
Step 1: This part of the eligibility analysis evaluates whether the claim falls within any statutory category. See MPEP 2106.03. Claim 1 recites: A learning apparatus for performing machine learning of a score function for binary classification, the learning apparatus comprising… Claim 1 is directed to an apparatus (Step 1: YES).
Step 2A prong 1: Does the claim recite a judicial exception? Claim 1 recites: specify the lowest one of the scores that have been calculated for the training data pieces to which the labels of positive examples have been added as a minimum score, and specify the highest one of the scores that have been calculated for the training data pieces to which the labels of negative examples have been added as a maximum score (identifying a lowest and highest data score is a mental process); select, from among the training data pieces to which the labels of positive examples have been added and from among the training data pieces to which the labels of negative examples have been added, training data pieces for which the calculated scores are equal to or higher than the minimum score and equal to or lower than the maximum score (selecting data within a range is a mental process), and generate a group of pairs of a positive example and a negative example from the selected training data pieces (grouping data into pairs is a mental process); and optimize and update a parameter of the score function through machine learning using hill climbing so as to, with regard to the generated group of pairs, increase the number of pairs in which a score of training data to which a label of a positive example has been added is higher than a score of training data to which a label of a negative example has been added (using hill climbing to update a parameter of a score function is a mathematical calculation)… select, from among the generated pairs, pairs in which a score of training data to which a label of a positive example has been added is lower than a score of training data to which a label of a negative example has been added (selecting pairs of data that fulfill a criteria is a mental process); ultimately select a set number of pairs randomly from among the selected pairs (selecting pairs randomly is a mental process); and generate the group of pairs composed of the ultimately selected pairs (grouping selected pairs is a mental process). These steps can be performed mentally or are mathematical calculations (Step 2A prong 1: YES).
Step 2A prong 2: Does the claim recite additional elements? Do those additional elements, considered individually and in combination, integrate the judicial exception into a practical application? Claim 1 recites: at least one memory storing instructions; and at least one processor configured to execute the instructions to: calculate scores by inputting, to the score function, a plurality of training data pieces to which labels of positive examples or negative examples have been added… wherein the at least one processor is further configured to execute the instructions to… The learning apparatus consisting of at least one memory and at least one processor, is recited at a high level of generality, i.e., as a generic computer performing generic computer functions without significantly more. Calculating scores by inputting data into a score function is insignificant extra-solution activity of data gathering (Step 2A prong 2: NO).
Step 2B: These elements are recited at such a high level of generality that they fail to integrate the abstract idea into a practical application, since they provide nothing more than mere instructions to implement an abstract idea on a generic computer (MPEP 2106.05(f)) or only amount to data gathering or outputting without significantly more (MPEP 2106.05(g)). These limitations, taken either alone or in combination, fail to provide an inventive concept (Step 2B: NO). Thus, the claim is not patent eligible.
Regarding claim 2, it recites limitations which further narrow the abstract idea by specifying more details of the mental and mathematical process that occurs (Claim 2, selecting a set number of training data between score thresholds is a mental process, and selecting pairs of positive and negative examples from those training data is also a mental process).
Regarding claim 4, it is a method that recites similar limitations to the apparatus of claim 1 and is rejected on the same grounds – see above.
Regarding claim 5 it recites similar limitations to claim 2 and is rejected on the same grounds – see above.
Regarding claim 7, it is an apparatus that recites similar limitations to the apparatus of claim 1 and is rejected on the same grounds – see above.
Regarding claim 8, it recites similar limitations to claim 2 and is rejected on the same grounds – see above.
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.
Claim(s) 1, 2, 4, 5, 7 and 8 is/are rejected under 35 U.S.C. 103 as being unpatentable over Ramanath et al. (US 20200005134 A1), herein Ramanath, in view of Nichols et al. (US 20160306876 A1), herein Nichols, Yao et al. (US 20210133518 A1), herein Yao, Lanckriet (US 20070162406 A1), Gong et al. (US 9552549 B1), herein Gong, and Martyanov (US 20210200612 A1).
Regarding claim 1, Ramanath teaches: A learning apparatus for performing machine learning of a score function for binary classification, the learning apparatus comprising: at least one memory storing instructions; and at least one processor configured to execute the instructions to: calculate scores by inputting, to the score function, a plurality of training data pieces to which labels of positive examples or negative examples have been added (¶160, In some example embodiments, the search system 216 uses pointwise learning, also referred to as ranking by binary classification. This method involves training a binary classifier utilizing each example in the training set with their labels, and then grouping the examples from the same search session together and ranking them based on their scores)… select, from among the training data pieces to which the labels of positive examples have been added and from among the training data pieces to which the labels of negative examples have been added, training data pieces… and generate a group of pairs of a positive example and a negative example from the selected training data pieces (¶161, the search system 216 forms pairs of examples with positive and negative labels respectively from the same session); and optimize and update a parameter of the score function through machine learning… so as to, with regard to the generated group of pairs, increase the number of pairs in which a score of training data to which a label of a positive example has been added is higher than a score of training data to which a label of a negative example has been added (¶161, trains the neural network to maximize the difference of scores between the paired positive and negative examples)…
Ramanath fails to teach: specify the lowest one of the scores that have been calculated for the training data pieces to which the labels of positive examples have been added as a minimum score, and specify the highest one of the scores that have been calculated for the training data pieces to which the labels of negative examples have been added as a maximum score… ultimately select a set number of pairs… from among the selected pairs; and generate the group of pairs composed of the ultimately selected pairs
However, in the same field of endeavor, Nichols teaches: specify the lowest one of the scores that have been calculated for the training data pieces to which the labels of positive examples have been added as a minimum score, and specify the highest one of the scores that have been calculated for the training data pieces to which the labels of negative examples have been added as a maximum score (¶45, The list of test set items can be reversed in order and processed in order from smallest to largest score until an appropriate number of low scoring items with known positive answers to the question have been identified at which point a minimum threshold can be set at the model result for that question, which below the minimum threshold below there is considered to be definitively no detection of the information in the content – and – ¶44, The test set items, their model score, and expected result can also be retrieved from the database and ordered by model score in descending order. The list of test set items can be iterated through until an appropriate number of high scoring items with known negative answers to the question have been identified at which point a maximum threshold is set at the model result for that item)… ultimately select a set number of data… from among the selected data; and generate the group of data composed of the ultimately selected data (¶41, Then, the training set can be retrieved and separated into batches of predetermined size (e.g. the default batch size can be 10,000 items). Training can be performed sequentially on the batch items, at the end of which the data can be saved back in the database (or elsewhere) – Nichols does not discuss selecting pairs specifically, but does discuss separating data into batches of predefined size).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to set thresholds based on low scoring positive examples and high scoring negative examples, and select set numbers of training examples as disclosed by Nichols in the apparatus disclosed by Ramanath to define a range of uncertainty, and prevent overfitting (¶45, below the minimum threshold below there is considered to be definitively no detection of the information in the content – and – ¶44, a maximum threshold is set… the score for that would be the upper threshold and any item scored higher than that would be returned by the system with an answer of yes – and – ¶42, Such a cycle of training and testing can continue until the results (e.g. accuracy indicator) of the tests no longer improve, indicating that the model has begun to overfit (e.g. models which are trained so heavily on specific content that they do not generalize well to new content), at which point a saved model from an immediately preceding previous training batch can be retrieved and presented as the final trained model).
Ramanath in view of Nichols fails to teach: for which the calculated scores are equal to or higher than the minimum score and equal to or lower than the maximum score.
However, in the same field of endeavor, Yao teaches: for which the calculated scores are equal to or higher than the minimum score and equal to or lower than the maximum score (¶10, Hard examples, as used herein, may refer to example images that may have a higher probability of being a false positive or a false negative. In some examples, the hard examples may include both positive and negative hard examples).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to select training data that proves problematic as disclosed by Yao in the apparatus disclosed by Ramanath in view of Nichols to improve accuracy (¶12, techniques described herein thus enable significant improvements in accuracy).
Ramanath in view of Nichols and Yao fails to teach: using hill climbing…
However, in the same field of endeavor, Lanckriet teaches: using hill climbing… (¶78, Proper adjustment of the two loss function parameters, C+ and C-, in order to derive an optimal signature (i.e., linear classifier) may be carried out using a modified "hill-climbing" algorithm. This modified hill-climbing procedure automatically adjusts the C+ and C- parameters in an efficient manner so as to provide an optimized LE score).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to use hill climbing as the method for optimizing a parameter as disclosed by Lanckriet in the apparatus disclosed by Ramanath in view of Nichols and Yao to efficiently discover proper parameter values (¶79, Therefore, the optimization algorithm does a local search, referred to as "hill-climbing." This local search includes inspecting the neighborhood of the current best (C+, C-) combination for a better (C+, C-) combination, while also inspecting the neighborhoods of suboptimal (C+, C-) combinations analyzed so far in order to avoid getting stuck in a noisy local maxima).
Ramanath in view of Nichols, Yao and Lanckriet fails to teach: wherein the at least one processor is further configured to execute the instructions to: select, from among the generated pairs, pairs in which a score of training data to which a label of a positive example has been added is lower than a score of training data to which a label of a negative example has been added.
However, in the same field of endeavor, Gong teaches: wherein the at least one processor is further configured to execute the instructions to: select, from among the generated pairs, pairs in which a score of training data to which a label of a positive example has been added is lower than a score of training data to which a label of a negative example has been added (Col. 8, line 13, the semantic ranking loss function 125 may compare label score for each label that is identified as a positive label in the training examples to the label score for each label that is identified as a negative label in the training examples, to determine if the neural network 110 scored the positive labels higher than the negative labels. In instances where a positive label was scored higher than a negative label, the neural network trainer 120 may determine that no error was made – Gong teaches using negative examples that score higher than positive examples to identify errors).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to select scores where a positive example scores lower than a negative example as disclosed by Gong in the pairwise selection step that is performed by the system disclosed by Ramanath in view of Nichols, Yao and Lanckriet to train the neural network more effectively (Col. 7, line 39, The error determined using the semantic ranking loss function 125 may be used to train the neural network 110 in any suitable manner by the neural network trainer 120).
Ramanath in view of Nichols, Yao, Lanckriet, and Gong fails to teach: selecting randomly…
However, in the same field of endeavor, Martyanov teaches: selecting randomly (¶46, selecting at random a vector from a data object that is a negative example and a vector from a data object that is a positive example and providing the pair)…
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to select pairs randomly as disclosed by Martyanov in the apparatus disclosed by Ramanath in view of Nichols, Yao, Lanckriet and Gong to provide input data for training the network (¶46, providing the pair as an input to the artificial neural network…A plurality of these input pairs may be provided for each epoch and a plurality of epochs may be performed on the artificial neural network).
Regarding claim 2, Ramanath in view of Nichols, Lanckriet, Gong and Martyanov fails to teach: The learning apparatus according to claim 1, wherein the at least one processor is further configured to execute the instructions to: select, from among the training data pieces to which the labels of positive examples have been added and from among the training data pieces to which the labels of negative examples have been added, a set number of training data pieces for which the calculated scores are equal to or higher than the minimum score and equal to or lower than the maximum score; and generate the group of pairs composed of the set number of pairs from the selected training data pieces
However, in the same field of endeavor, Yao teaches: wherein the at least one processor is further configured to execute the instructions to: select, from among the training data pieces to which the labels of positive examples have been added and from among the training data pieces to which the labels of negative examples have been added, a set number of training data pieces for which the calculated scores are equal to or higher than the minimum score and equal to or lower than the maximum score; and generate the group of pairs composed of the set number of pairs from the selected training data pieces (¶10, Hard examples, as used herein, may refer to example images that may have a higher probability of being a false positive or a false negative. In some examples, the hard examples may include both positive and negative hard examples – and – ¶62, For example, the multi-scale hard example miner module 712 may be configured to receive a plurality of candidate boxes and select a fixed number of high-loss examples from the plurality of candidate boxes to be used to execute a back-propagation and fine-tune parameters of the detection network).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to select a set number of training examples that proves problematic as disclosed by Yao in the apparatus disclosed by Ramanath in view of Nichols, Lanckriet, Gong and Martyanov to improve accuracy while reducing the training data set size (¶12, techniques described herein thus enable significant improvements in accuracy – and – ¶46, In some examples, using hard examples of both positive examples and negative examples may improve detection accuracy, while also reducing the number of example images to be used to train a CNN).
Regarding claim 4, it is a method that recites similar limitations to claim 1 and is rejected on the same grounds – see above.
Regarding claim 5, it recites similar limitations to claim 2 and is rejected on the same grounds – see above.
Regarding claim 7, it is an apparatus that recites similar limitations to claim 1 and is rejected on the same grounds – see above.
Regarding claim 8, it recites similar limitations to claim 2 and is rejected on the same grounds – see above.
Conclusion
Applicant's amendment necessitated the new ground(s) of rejection presented in this Office action. Accordingly, THIS ACTION IS MADE FINAL. See MPEP § 706.07(a). 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 HARRISON CHAN YOUNG KIM whose telephone number is (571)272-0713. The examiner can normally be reached Monday - Friday 9:00 am - 5:00 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, Cesar Paula can be reached at (571) 272-4128. 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.
/HARRISON C KIM/ Examiner, Art Unit 2145
/CESAR B PAULA/ Supervisory Patent Examiner, Art Unit 2145
Prosecution Insights