DETAILED ACTION
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 responsive to communications: Amendment filed on 4/30/2026.
Claims 1-20 are pending. Claims 1, 10, and 15 are independent.
The previous rejection of claims 1-20 under 35 USC § 103 have been withdrawn in view of the amendment.
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, 5-7, 10-11, and 14-18 is/are rejected under 35 U.S.C. 103 as being unpatentable over Mohseni et al. (US2021/0142160) in view of Zhang ‘968 (US2020/0073968) and Lu et al. (“An Embarrassingly Simple Approach to Visual Domain Adaptation”).
In regards to claim 1, Mohseni et al. discloses a method comprising:
receiving input data for inputting into an AI (artificial intelligence) based system, (Mohseni et al. fig. 5 502 para[0070], at block 502 receives input data);
generating probability distribution functions, in the in-domain feature space and the out-of-domain feature space, for the input data and for the data that the AI based system is trained to classify (Mohseni et al. fig. 5 506para[0070], uses softmax function to generate probability distribution) and
determining whether the input data is out-of-domain of the AI based system based on the probability distribution functions for the input data and for the data that the AI based system is trained to classify (Mohseni et al. fig. 5 510 para[0071], determines if input data is out-of-distribution using values of softmax probability distribution).
Mohseni further discloses modeling and in-domain feature space of the AI based system and an out-of-domain feature space of the AI based system, the in-domain feature space corresponding to features of data that the AI based system is trained to classify and the out-of-domain feature space corresponding to features of data that the AI based system is not trained to classify (Mohseni et al. fig. 5 504 para[0070], generating values at a first set of output nodes, and second set of output nodes, based on received input data, and generating softmax values corresponding to output nodes to determine if input was out-of-distribution or not).
Mohseni et al. does not explicitly disclose the AI based system comprising a machine learning based network comprising one or more projection layers implemented with a projection matrix;
Modeling 1) an in-domain feature space based on the projection matrix and 2) and out of domain feature space based on the transpose of the feature space.
However Zhang ‘968 discloses the AI based system comprising a machine learning based network comprising one or more projection layers implemented with a projection matrix (Zhang ‘968 para[0043], the attention model can be configured to identify the core regions of the sketches and synthetic sketches during the learning process. The attention model can include a convolutional layer and a softmax function with a threshold can be applied to the output for obtaining a binary attention mask);
modelling 1) an in-domain feature space of the AI based system based on the projection matrix and 2) an out-of-domain feature space of the AI based system based on a transpose of the projection matrix, the in-domain feature space corresponding to features of data that the AI based system is trained to classify and the out-of-domain feature space corresponding to features of data that the AI based system is not trained to classify (Zhang ‘968 para[0110]-[0120], models feature space based on projection matrix D and transposed matrix of D).
It would have been obvious to one of ordinary skill in the art before the filing date of the invention to have combined the out of distribution detection of Mohseni et al. with the domain migration method of Zhang ‘968 in order to preserve semantic consistency between domains (Zhang ‘968 para[0010]).
Mohseni et al. does not explicitly disclose constructing a second machine learning based network having weights defined based on the transpose of the projection matrix;
by fitting parameters of the probability distribution functions to in-domain features computed by the first machine learning based network and to out-of-domain features computed by the second machine learning based network.
However Lu et al. discloses constructing a second machine learning based network having weights defined based on the transpose of the projection matrix; (Lu et al. pg3406 section III.A para1-2, consider the difference between class I and the mean of all classes except I (denoted by \i));
by fitting parameters of the probability distribution functions to in-domain features computed by the first machine learning based network and to out-of-domain features computed by the second machine learning based network (Lu et al. fig. 5 pg3407 section III.C para4-5, iteratively fit parameters for in-domain (target) and out-of-domain (source) machine learning networks).
It would have been obvious to one of ordinary skill in the art before the filing date of the invention to have combined to out of distribution detection of Mohseni et al. with the domain adaptation method of Lu et al. in order to efficiently generate a class specific projection (Lu et al. pg3404 section1 para4).
In regards to claim 2, Mohseni et al. as modified by Zhang ‘968 and Lu et al. discloses the method of claim 1, wherein modelling an in-domain feature space of the AI based system and an out-of-domain feature space of the AI based system comprises:
computing the in-domain feature space based on the projection matrix (Zhang ‘968 para[0042], identifies in-domain feature space of image and sketch domains);
computing one or more orthogonal linear projection matrices for the out-of-domain feature space based on the transpose of the projection matrix (Zhang ‘968 para [0054], uses discriminator to identify out of domain space); and
computing the out-of-domain feature space based on the one or more orthogonal linear projection matrices for the out-of-domain feature space (Zhang ‘968 para[0062], computes out of domain feature space based on generator and discriminator).
In regards to claim 5, Mohseni et al. as modified by Zhang et al. and Lu et al. discloses the method of claim 1, further comprising:
in response to determining that the input data is out-of-domain of the AI based system:
annotating the input data (Mohseni et al. para[0060], assigns pseudo-labels with values of 100 or greater for OOD inputs); and
training the AI based system based on the annotated input data (Mohseni et al. para[0060], pseudo labeled samples used to train neural network).
In regards to claim 6, Mohseni et al. as modified by Zhang et al. and Lu et al. discloses the method of claim 1, further comprising:
in response to determining that the input data is not out-of-domain of the AI based system, generating a prediction from the input data by the AI based system (Mohseni et al. para[0106], infer or predict information based on input).
In regards to claim 7, Mohseni et al. as modified by Zhang et al. and Lu et al. discloses the method of claim 1, wherein the receiving, the modelling, the generating, and the determining are performed by a module combined with the AI based system and the module combined with the AI based system generates a prediction from the input data and the determination of whether the input data is out-of-domain of the AI based system (Mohseni et al. para[0109], infers OOD input data)
Claim 10-11 and 14 recite substantially similar limitations to claims 1-2 and 7. Thus claims 10-11 and 14 are rejected along the same rationale as claims 1-2 and 7.
Claims 15-16 and 17-18 recite substantially similar limitations to claims 1-2, and 5-6. Thus claims 15-16 and 17-18 are rejected along the same rationale as claims 1-2, and 5-6.
Claim(s) 3-4, 8-9, 12-13, and 19-20 is/are rejected under 35 U.S.C. 103 as being unpatentable over Mohseni et al. in view of Zhang ‘968, Lu et al. and Mohseni ‘597 (US2022/0318557).
In regards to claim 3, Mohseni et al. as modified by Zhang ‘968 and Lu et al. discloses the method of claim 1. Mohseni et al. does not explicitly disclose wherein generating probability distribution functions, in the in-domain feature space and the out-of-domain feature space, for the input data and for the data that the AI based system is trained to classify comprises:
generating the probability distribution functions using a Gaussian process model or a combination of Gaussian probability distribution models fitted to available data.
However Mohseni ‘557 discloses wherein generating probability distribution functions, in the in-domain feature space and the out-of-domain feature space, for the input data and for the data that the AI based system is trained to classify comprises:
generating the probability distribution functions using a Gaussian process model or a combination of Gaussian probability distribution models fitted to available data (Mohseni ‘557 para[0058], transforms input data using Gaussian process).
It would have been obvious to one of ordinary skill in the art before the filing date of the invention to have combined the out of distribution detection of Mohseni et al. with the out-of-distribution identification method of Mohseni ‘557 in order to identify and prevent classification errors (Mohseni ‘557 para[0002]).
In regards to claim 4, Mohseni et al. as modified by Zhang ‘968 and Lu et al. discloses the method of claim 1. Mohseni et al. does not explicitly disclose further comprising:
in response to determining that the input data is out-of-domain of the AI based system, transmitting a notification to a user for reviewing a prediction generated by the AI based system from the input data.
However Mohseni ‘557 discloses further comprising:
in response to determining that the input data is out-of-domain of the AI based system, transmitting a notification to a user for reviewing a prediction generated by the AI based system from the input data (Mohseni ‘557 fig. 6 para[0088], generates OOD notification if OOD data is detected at block 606).
It would have been obvious to one of ordinary skill in the art before the filing date of the invention to have combined the out of distribution detection of Mohseni et al. with the out-of-distribution identification method of Mohseni ‘557 in order to identify and prevent classification errors (Mohseni ‘557 para[0002]).
In regards to claim 8, Mohseni et al. as modified by Zhang ‘968 and Lu et al. discloses the method of claim 1. Mohseni et al. does not explicitly disclose further comprising:
selecting one of a plurality of algorithms of the AI based system based on the determining.
However Mohseni ‘557 discloses further comprising:
selecting one of a plurality of algorithms of the AI based system based on the determining (Mohseni ‘557 para[0199], selects from a variety of vision algorithms for object detection).
It would have been obvious to one of ordinary skill in the art before the filing date of the invention to have combined the out of distribution detection of Mohseni et al. with the out-of-distribution identification method of Mohseni ‘557 in order to identify and prevent classification errors (Mohseni ‘557 para[0002]).
In regards to claim 9, Mohseni et al. as modified by Zhang ‘968 and Lu et al. discloses the method of claim 1. Mohseni et al. does not explicitly disclose wherein the AI based system is for medical imaging analysis.
However Mohseni ‘557 discloses wherein the AI based system is for medical imaging analysis (Mohseni ‘557 para[0078], domain of application includes analysis of medical imagery such as X-ray imagery).
It would have been obvious to one of ordinary skill in the art before the filing date of the invention to have combined the out of distribution detection of Mohseni et al. with the out-of-distribution identification method of Mohseni ‘557 in order to identify and prevent classification errors (Mohseni ‘557 para[0002]).
Claims 12-13 recite substantially similar limitation to claims 3-4. Thus claims 12-13 are rejected along the same rationale as claims 3-4.
Claims 19-20 recite substantially similar limitations to claims 8-9. Thus claims 19-20 are rejected along the same rationale as claims 8-9.
Response to Arguments
Applicant’s arguments with respect to claims 1-20 have been considered but are moot because the arguments do not apply the current rejection.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure. Tan et al. (US20210034965) teaches generating an in-domain and out-of-domain classifiers.
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 NICHOLAS HASTY whose telephone number is (571)270-7775. The examiner can normally be reached Monday-Friday 8:30am-5:00pm.
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, Matt Ell can be reached at (571)270-3264. 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.
/N.H/Examiner, Art Unit 2141
/MATTHEW ELL/Supervisory Patent Examiner, Art Unit 2141
Prosecution Insights