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 .
Claims 1-16 are pending for examination. Claims 1, 13, and 14 are independent.
Claim Interpretation
The following is a quotation of 35 U.S.C. 112(f):
(f) Element in Claim for a Combination. – An element in a claim for a combination may be expressed as a means or step for performing a specified function without the recital of structure, material, or acts in support thereof, and such claim shall be construed to cover the corresponding structure, material, or acts described in the specification and equivalents thereof.
The following is a quotation of pre-AIA 35 U.S.C. 112, sixth paragraph:
An element in a claim for a combination may be expressed as a means or step for performing a specified function without the recital of structure, material, or acts in support thereof, and such claim shall be construed to cover the corresponding structure, material, or acts described in the specification and equivalents thereof.
The claims in this application are given their broadest reasonable interpretation using the plain meaning of the claim language in light of the specification as it would be understood by one of ordinary skill in the art. The broadest reasonable interpretation of a claim element (also commonly referred to as a claim limitation) is limited by the description in the specification when 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, is invoked.
As explained in MPEP § 2181, subsection I, claim limitations that meet the following three-prong test will be interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph:
(A) the claim limitation uses the term “means” or “step” or a term used as a substitute for “means” that is a generic placeholder (also called a nonce term or a non-structural term having no specific structural meaning) for performing the claimed function;
(B) the term “means” or “step” or the generic placeholder is modified by functional language, typically, but not always linked by the transition word “for” (e.g., “means for”) or another linking word or phrase, such as “configured to” or “so that”; and
(C) the term “means” or “step” or the generic placeholder is not modified by sufficient structure, material, or acts for performing the claimed function.
Use of the word “means” (or “step”) in a claim with functional language creates a rebuttable presumption that the claim limitation is to be treated in accordance with 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph. The presumption that the claim limitation is interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, is rebutted when the claim limitation recites sufficient structure, material, or acts to entirely perform the recited function.
Absence of the word “means” (or “step”) in a claim creates a rebuttable presumption that the claim limitation is not to be treated in accordance with 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph. The presumption that the claim limitation is not interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, is rebutted when the claim limitation recites function without reciting sufficient structure, material or acts to entirely perform the recited function.
Claim limitations in this application that use the word “means” (or “step”) are being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, except as otherwise indicated in an Office action. Conversely, claim limitations in this application that do not use the word “means” (or “step”) are not being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, except as otherwise indicated in an Office action.
This application includes one or more claim limitations that do not use the word “means,” but are nonetheless being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, because the claim limitation(s) uses a generic placeholder that is coupled with functional language without reciting sufficient structure to perform the recited function and the generic placeholder is not preceded by a structural modifier. Such claim limitation(s) is/are:
“a first model acquisition unit configured to” in claim 13.
“a second model acquisition unit configured to” in claim 13.
“a data acquisition unit configured to” in claim 13.
“a first inference unit configured to” in claim 13.
“a second inference unit configured to” in claim 13.
“a generation unit configured to” in claim 13.
Examiner finds structure for the different units in para 0070 of the specification.
Because this/these claim limitation(s) is/are being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, it/they is/are being interpreted to cover the corresponding structure described in the specification as performing the claimed function, and equivalents thereof.
If applicant does not intend to have this/these limitation(s) interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, applicant may: (1) amend the claim limitation(s) to avoid it/them being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph (e.g., by reciting sufficient structure to perform the claimed function); or (2) present a sufficient showing that the claim limitation(s) recite(s) sufficient structure to perform the claimed function so as to avoid it/them being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph.
Claim Rejections - 35 USC § 101
35 U.S.C. 101 reads as follows:
Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.
Claims 1-16 rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
Step 1
According to the first part of the analysis, in the instant case, each of the claims falls within one of the four statutory categories (i.e., process, machine, manufacture, or composition of matter).
Regarding Claim 1:
2A Prong 1:
generating a third correct answer label for the third training data based on a match between the first inference result and the second inference result. (This step for generating an answer label is practically performable in the human mind and is understood to be a recitation of a mental process (i.e., judgment/evaluation).)
2A Prong 2: This judicial exception is not integrated into a practical application.
Additional elements:
A label generation method in which a computer executes steps of: (The computer recited is understood to be mere instructions to implement an abstract idea on a computer, or merely uses a computer as a tool to perform an abstract idea - see MPEP 2106.05(f))
acquiring a trained first inference model generated by machine learning using a plurality of first data sets each configured by a combination of first training data in a source domain and a first correct answer label indicating a correct answer of an inference task for the first training data; (This step is directed to receiving information, which is understood to be insignificant extra-solution activity and data gathering. See MPEP 2106.05(g).)
acquiring a trained second inference model generated by machine learning using a plurality of second data sets each configured by a combination of second training data generated by applying a disturbance to the first training data and a second correct answer label indicating a correct answer of the inference task for the second training data; (This step is directed to receiving information, which is understood to be insignificant extra-solution activity and data gathering. See MPEP 2106.05(g).)
acquiring third training data; (This step is directed to receiving information, which is understood to be insignificant extra-solution activity and data gathering. See MPEP 2106.05(g).)
acquiring, using the trained first inference model, a first inference result obtained by performing the inference task on the acquired third training data; (This step is directed to receiving information, which is understood to be insignificant extra-solution activity and data gathering. See MPEP 2106.05(g).)
acquiring, using the trained second inference model, a second inference result obtained by performing the inference task on the acquired third training data; (This step is directed to receiving information, which is understood to be insignificant extra-solution activity and data gathering. See MPEP 2106.05(g).)
The additional elements as disclosed above alone or in combination do not integrate the judicial exception into practical application as they are insignificant extra solution activity in combination of generic computer functions that are implemented to perform the disclosed abstract idea above.
2B: The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception.
Additional elements:
A label generation method in which a computer executes steps of: (The computer recited is understood to be mere instructions to implement an abstract idea on a computer, or merely uses a computer as a tool to perform an abstract idea - see MPEP 2106.05(f))
acquiring a trained first inference model generated by machine learning using a plurality of first data sets each configured by a combination of first training data in a source domain and a first correct answer label indicating a correct answer of an inference task for the first training data; (This step is directed to receiving information, which is understood to be insignificant extra-solution activity and is well understood, routine and conventional activity of transmitting and receiving data as identified by the court (MPEP2106.05(d)(ll)(i))))
acquiring a trained second inference model generated by machine learning using a plurality of second data sets each configured by a combination of second training data generated by applying a disturbance to the first training data and a second correct answer label indicating a correct answer of the inference task for the second training data; (This step is directed to receiving information, which is understood to be insignificant extra-solution activity and is well understood, routine and conventional activity of transmitting and receiving data as identified by the court (MPEP2106.05(d)(ll)(i))))
acquiring third training data; (This step is directed to receiving information, which is understood to be insignificant extra-solution activity and is well understood, routine and conventional activity of transmitting and receiving data as identified by the court (MPEP2106.05(d)(ll)(i))))
acquiring, using the trained first inference model, a first inference result obtained by performing the inference task on the acquired third training data; (This step is directed to receiving information, which is understood to be insignificant extra-solution activity and is well understood, routine and conventional activity of transmitting and receiving data as identified by the court (MPEP2106.05(d)(ll)(i))))
acquiring, using the trained second inference model, a second inference result obtained by performing the inference task on the acquired third training data; (This step is directed to receiving information, which is understood to be insignificant extra-solution activity and is well understood, routine and conventional activity of transmitting and receiving data as identified by the court (MPEP2106.05(d)(ll)(i))))
The additional elements as disclosed above in combination of the abstract idea
are not sufficient to amount to significantly more than the judicial exception as they are
well, understood, routine and conventional activity as disclosed in combination
of generic computer functions that are implemented to perform the disclosed abstract idea above.
Regarding Claim 13: see the rejection of claim 1 above. Same rationale applies.
2A Prong 2 & 2B: The claim recites another additional element “A label generation device comprising: A label generation device comprising: a first model acquisition unit configured to […] a second model acquisition unit configured to […] a data acquisition unit configured to […] a first inference unit configured to […] a second inference unit configured to […] a generation unit configured to” (These limitations are understood to be mere instructions to apply the exception using generic computer components - see MPEP 2106.05(f))
Regarding Claim 14: see the rejection of claim 1 above. Same rationale applies.
2A Prong 2 & 2B: The claim recites another additional element “A non-transitory computer readable medium storing a label generation program, the label generation program configured to cause a computer to execute steps of:” (mere instructions to apply the exception using a generic computer component - see MPEP 2106.05(f))
Regarding Claim 2
2A Prong 1: The claim does not recite any Abstract idea.
2A Prong 2:
wherein the third training data is acquired in a target domain different from the source domain. (This step is directed to receiving information, which is understood to be insignificant extra-solution activity and data gathering. See MPEP 2106.05(g).)
2B:
wherein the third training data is acquired in a target domain different from the source domain. (This step is directed to receiving information, which is understood to be insignificant extra-solution activity and is well understood, routine and conventional activity of transmitting and receiving data as identified by the court (MPEP2106.05(d)(ll)(i))))
Regarding Claim 3
2A Prong 1:
the applying a disturbance to the first training data is configured by transforming the first training data using a trained transformation model, and the trained transformation model is generated to acquire an ability to transform a style of the first training data into a style of the third training data (This step is practically performable in the human mind and is understood to be a recitation of a mental process (i.e., judgment/evaluation).)
2A Prong 2 & 2B:
by machine learning (This step is adding the words “apply it” (or an equivalent) with the judicial exception, or merely applying machine learning as a tool to perform the abstract idea (i.e., transforming a style) - see MPEP 2106.05(f).)
Regarding Claim 4
2A Prong 1: The claim does not recite any Abstract idea.
2A Prong 2 & 2B:
wherein the first inference model and the second inference model are further trained by adversarial learning with an identification model, and
the adversarial learning includes:
training the identification model using the first training data and the third training data to identify which training data of the first training data and the third training data an inference result of the first inference model is for;
training the first inference model using the first training data and the third training data to degrade identification performance of the identification model;
training the identification model using the second training data and the third training data to identify which training data of the second training data and the third training data an inference result of the second inference model is for; and
training the second inference model using the second training data and the third training data to degrade identification performance of the identification model. (Training a machine learning model is understood as mere instructions to implement an abstract idea (e.g., generate inferences) on a computer - see MPEP 2106.05(f).))
Regarding Claim 5
2A Prong 1: The claim does not recite any Abstract idea.
2A Prong 2 & 2B:
wherein the computer further executes a step of outputting the generated third correct answer label. (This step is directed to transmitting or receiving information, which is understood to be insignificant extra-solution activity and data gathering. See MPEP 2106.05(g).)
2B:
wherein the computer further executes a step of outputting the generated third correct answer label. (This step is directed to receiving information, which is understood to be insignificant extra-solution activity and is well understood, routine and conventional activity of transmitting and receiving data as identified by the court (MPEP2106.05(d)(ll)(i))))
Regarding Claim 6
2A Prong 1:
wherein
the inference task is extracting a region including a feature, and
the generating the third correct answer label based on the match includes:
specifying an overlapping portion of a region extracted as the first inference result and a region extracted as the second inference result (This step is practically performable in the human mind and is understood to be a recitation of a mental process (i.e., judgment/evaluation).); and
generating the third correct answer label so as to indicate the overlapping portion as a correct answer of the inference task in a case where a size of the specified overlapping portion exceeds a threshold. (This step is practically performable in the human mind and is understood to be a recitation of a mental process (i.e., judgment/evaluation).)
2A Prong 2 & 2B: The claim does not recite any additional elements.
Regarding Claim 7
2A Prong 1:
wherein the inference task is identifying a class of a feature included in data, and the generating the third correct answer label based on the match includes generating the third correct answer label so as to, in a case where a class identified as the first inference result and a class identified as the second inference result match, indicate the matched class. (These steps are practically performable in the human mind and is understood to be a recitation of a mental process (i.e., judgment/evaluation).)
2A Prong 2 & 2B: The claim does not recite any additional elements.
Regarding Claim 8
2A Prong 1:
wherein the inference task includes at least one of extracting a region including a feature in the image data and identifying a class of a feature included in the image data. (This step is practically performable in the human mind and is understood to be a recitation of a mental process (i.e., judgment/evaluation).)
2A Prong 2 & 2B:
each of the training data includes image data, (The specification of data to be stored is understood to be a field of use limitation. The limitation further specifies the training data - See MPEP 2106.05(h).)
Regarding Claim 9
2A Prong 1:
the inference task includes extracting a region including a feature in the image data, (This step is practically performable in the human mind and is understood to be a recitation of a mental process (i.e., judgment/evaluation).)
2A Prong 2 & 2B:
wherein each of the training data includes image data, (The specification of data to be stored is understood to be a field of use limitation. The limitation further specifies the training data - See MPEP 2106.05(h).)
the first inference model and the second inference model are further trained by adversarial learning with an identification model, and
the adversarial learning includes:
training the identification model using the first training data and the third training data to identify for each pixel which training data of the first training data and the third training data an inference result of the first inference model is for;
training the first inference model using the first training data and the third training data to degrade identification performance of the identification model;
training the identification model using the second training data and the third training data to identify for each pixel which training data of the second training data and the third training data an inference result of the second inference model is for; and
training the second inference model using the second training data and the third training data to degrade identification performance of the identification model. (Training a machine learning model is understood as mere instructions to implement an abstract idea (e.g., generate inferences) on a computer - see MPEP 2106.05(f).))
Regarding Claim 10
2A Prong 1:
the inference task includes at least one of extracting a region including a feature in the sound data and identifying a class of a feature included in the sound data. (This step is practically performable in the human mind and is understood to be a recitation of a mental process (i.e., judgment/evaluation).)
2A Prong 2 & 2B:
wherein each of the training data includes sound data, (The specification of data to be stored is understood to be a field of use limitation. The limitation further specifies the training data - See MPEP 2106.05(h).)
Regarding Claim 11
2A Prong 1:
the inference task includes at least one of extracting a region including a feature in the sensing data and identifying a class of a feature included in the sensing data. (This step is practically performable in the human mind and is understood to be a recitation of a mental process (i.e., judgment/evaluation).)
2A Prong 2 & 2B:
wherein each of the training data includes sensing data, (The specification of data to be stored is understood to be a field of use limitation. The limitation further specifies the training data - See MPEP 2106.05(h).)
Regarding Claim 12
2A Prong 1: The claim does not recite any Abstract idea.
2A Prong 2:
A model generation method in which a computer executes steps of:
acquiring a plurality of third data sets generated by associating the third correct answer label generated by the label generation method according to claim 1 with the third training data; (This step is directed to receiving information, which is understood to be insignificant extra-solution activity and data gathering. See MPEP 2106.05(g).) and
performing machine learning of a third inference model by using the plurality of acquired third data sets, the machine learning being configured by training, for each of the third data sets, the third inference model such that an inference result obtained by performing the inference task by the third inference model on the third training data conforms to a correct answer indicated by the third correct answer label. (This step is adding the words “apply it” (or an equivalent) with the judicial exception or merely applying machine learning as a tool to perform the abstract idea (i.e., predicting) - see MPEP 2106.05(f).)
2B:
acquiring a plurality of third data sets generated by associating the third correct answer label generated by the label generation method according to claim 1 with the third training data; (This step is directed to transmitting or receiving information, which is understood to be insignificant extra-solution activity and is well understood, routine and conventional activity of transmitting and receiving data as identified by the court (MPEP2106.05(d)(ll)(i)))) and
performing machine learning of a third inference model by using the plurality of acquired third data sets, the machine learning being configured by training, for each of the third data sets, the third inference model such that an inference result obtained by performing the inference task by the third inference model on the third training data conforms to a correct answer indicated by the third correct answer label. (This step is adding the words “apply it” (or an equivalent) with the judicial exception or merely applying machine learning as a tool to perform the abstract idea (i.e., predicting) - see MPEP 2106.05(f).)
Regarding Claim 15
2A Prong 1: The claim does not recite any Abstract idea.
2A Prong 2:
A model generation device comprising:
a data acquisition unit configured to acquire a plurality of third data sets generated by associating the third correct answer label generated by the label generation method according to claim 1 with the third training data; (This step is directed to receiving information, which is understood to be insignificant extra-solution activity and data gathering. See MPEP 2106.05(g).) and
a learning processing unit configured to perform machine learning of a third inference model by using the plurality of acquired third data sets, the machine learning being configured by training, for each of the third data sets, the third inference model such that an inference result obtained by performing the inference task by the third inference model on the third training data conforms to a correct answer indicated by the third correct answer label. (This step is adding the words “apply it” (or an equivalent) with the judicial exception or merely applying machine learning as a tool to perform the abstract idea (i.e., predicting) - see MPEP 2106.05(f).)
2B:
a data acquisition unit configured to acquire a plurality of third data sets generated by associating the third correct answer label generated by the label generation method according to claim 1 with the third training data; (This step is directed to transmitting or receiving information, which is understood to be insignificant extra-solution activity and is well understood, routine and conventional activity of transmitting and receiving data as identified by the court (MPEP2106.05(d)(ll)(i)))) and
a learning processing unit configured to perform machine learning of a third inference model by using the plurality of acquired third data sets, the machine learning being configured by training, for each of the third data sets, the third inference model such that an inference result obtained by performing the inference task by the third inference model on the third training data conforms to a correct answer indicated by the third correct answer label. (This step is adding the words “apply it” (or an equivalent) with the judicial exception or merely applying machine learning as a tool to perform the abstract idea (i.e., predicting) - see MPEP 2106.05(f).)
Regarding Claim 16
2A Prong 1: The claim does not recite any Abstract idea.
2A Prong 2:
A non-transitory computer readable medium storing a model generation program, the model generation program configured to cause a computer to execute steps of: (mere instructions to apply the exception using a generic computer component - see MPEP 2106.05(f))
acquiring a plurality of third data sets generated by associating the third correct answer label generated by the label generation method according to claim 1with the third training data; (This step is directed to receiving information, which is understood to be insignificant extra-solution activity and data gathering. See MPEP 2106.05(g).) and
performing machine learning of a third inference model by using the plurality of acquired third data sets, the machine learning being configured by training, for each of the third data sets, the third inference model such that an inference result obtained by performing the inference task by the third inference model on the third training data conforms to a correct answer indicated by the third correct answer label. (This step is adding the words “apply it” (or an equivalent) with the judicial exception or merely applying machine learning as a tool to perform the abstract idea (i.e., predicting) - see MPEP 2106.05(f).)
2B:
A non-transitory computer readable medium storing a model generation program, the model generation program configured to cause a computer to execute steps of: (mere instructions to apply the exception using a generic computer component - see MPEP 2106.05(f))
acquiring a plurality of third data sets generated by associating the third correct answer label generated by the label generation method according to claim 1with the third training data; (This step is directed to transmitting or receiving information, which is understood to be insignificant extra-solution activity and is well understood, routine and conventional activity of transmitting and receiving data as identified by the court (MPEP2106.05(d)(ll)(i)))) and
performing machine learning of a third inference model by using the plurality of acquired third data sets, the machine learning being configured by training, for each of the third data sets, the third inference model such that an inference result obtained by performing the inference task by the third inference model on the third training data conforms to a correct answer indicated by the third correct answer label. (This step is adding the words “apply it” (or an equivalent) with the judicial exception or merely applying machine learning as a tool to perform the abstract idea (i.e., predicting) - see 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 text of those sections of Title 35, U.S. Code not included in this action can be found in a prior Office action.
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.
Claim(s) 1, 5-8, and 12-16 is/are rejected under 35 U.S.C. 103 as being unpatentable over Luong et al. (US 20200334538 A1, hereinafter "Luong") in view of Chen et al. ("Semi-Supervised Semantic Segmentation with Cross Pseudo Supervision", hereinafter Chen).
Regarding Claim 1
Luong discloses: A label generation method in which a computer executes steps of ([Abstract, Para 0110-0115, and Fig 9] describes a computer for training models and generating pseudo-labels.):
acquiring a trained first inference model generated by machine learning using a plurality of first data sets each configured by a combination of first training data in a source domain and a first correct answer label indicating a correct answer of an inference task for the first training data; ([Para 0037-0038, 0050, 0066, 0080 and Fig 1-3] describes a trained first machine learning model (i.e. trained first inference model) that represents a teacher model (i.e. source domain) trained on labeled data (i.e. first data set).)
acquiring a trained second inference model generated by machine learning using a plurality of second data sets each configured by a combination of second training data generated by applying a disturbance to the first training data and a second correct answer label indicating a correct answer of the inference task for the second training data; ([Para 0037-0039 and Fig 1-3] describes a trained second machine learning model (i.e. trained second inference model) that represents a student model trained on injected noise to the dataset (i.e. second training data).)
acquiring third training data; ([Para 0036, 0051, 0067, and Fig 1-3] describes unlabeled data 150 (i.e. third training data).)
acquiring, using the trained first inference model, a first inference result obtained by performing the inference task on the acquired third training data; ([Para 0051 and Fig 1-3] describes generating pseudo label (i.e. first inference result) by processing unlabeled data 150 (i.e. inference task on the acquired third training data) using the teacher model (i.e. trained first inference model).);
acquiring, using the trained second inference model, a second inference result obtained by performing the inference task on the acquired third training data ([Para 0057 and Fig 1-3] describes generating pseudo label (i.e. second inference result) by processing unlabeled data 150 (i.e. inference task on the acquired third training data) using the trained/learned second machine-learning model (i.e. trained second inference model).); and
generating a third correct answer label for the third training data ([Para 0051, 0057, and Fig 1-3] describes generating pseudo label (i.e. third correct answer label) by processing unlabeled data 150 (i.e. third training data).)
Luong does not explicitly disclose: generating a third correct answer label for the third training data based on a match between the first inference result and the second inference result.
However, Chen discloses in the same field of endeavor: generating a third correct answer label for the third training data based on a match between the first inference result and the second inference result. ([Abstract, Section 3, and Fig 1(a)] describes generating a pseudo label map (i.e. third correct answer) using expanded unlabeled data (i.e. third training data) based on cross pseudo supervision similarity (i.e. match) between the predictions of two perturbed networks (i.e. first and second inference results) for the same input image.)
It would have been obvious to a person of ordinary skill in art before the effective filling date of the invention to implement the function of Cross pseudo supervision disclosed by Chen into the method of Teacher Student learning disclosed by Luong to generate an answer label that match results between first and second models. The modification would have been obvious because one of the ordinary skills of the art would be motivated to utilize the feature of Cross pseudo supervision disclosed by Chen as all the references are in the field of machine learning. A person of ordinary skill of the art would have been motivated to perform the combination for being able to obtain high similarity results between the predictions of two perturbed networks.
Regarding Claim 13
Luong in view of Chen discloses: A label generation device comprising: a first model acquisition unit configured to […] a second model acquisition unit configured to […] a data acquisition unit configured to […] a first inference unit configured to […] a second inference unit configured to […] a generation unit configured to ([Abstract, Para 0110-0115, Fig 1, and Fig 9], Luong describes a computer with different units for training models and generating pseudo-labels.)
(Claim 13 is a device claim that corresponds to claim 1 and the rest of the limitations are rejected on the same ground)
Regarding Claim 14
Luong in view of Chen discloses: A non-transitory computer readable medium storing a label generation program, the label generation program configured to cause a computer ([Abstract, Para 0110-0115, and Fig 9], Luong describes a computer for training models and generating pseudo-labels.) to execute steps of: (Claim 14 is a non-transitory computer readable medium claim that corresponds to claim 1 and the rest of the limitations are rejected on the same ground)
Regarding Claim 5
Luong in view of Chen discloses: The label generation method according to claim 1, wherein the computer further executes a step of outputting the generated third correct answer label. ([Para 0002, 0112], Luong describes outputting results.)
Regarding Claim 6
Luong in view of Chen discloses: The label generation method according to claim 1, wherein
the inference task is extracting a region including a feature, and
the generating the third correct answer label based on the match includes:
specifying an overlapping portion of a region extracted as the first inference result and a region extracted as the second inference result; and
generating the third correct answer label so as to indicate the overlapping portion as a correct answer of the inference task in a case where a size of the specified overlapping portion exceeds a threshold. ([Section 3, Section B. Network Perturbation, and Fig 1(a)], Chen describes cross pseudo supervision similarity and overlap between segmentation predictions of images.)
Regarding Claim 7
Luong in view of Chen discloses: The label generation method according to claim 1, wherein
the inference task is identifying a class of a feature included in data ([Para 0016, 0048-0055], Luong describes classifying image data.), and
the generating the third correct answer label based on the match includes generating the third correct answer label so as to, in a case where a class identified as the first inference result and a class identified as the second inference result match, indicate the matched class. ([Abstract, Section 3, and Fig 1(a)], Chen describes cross pseudo supervision similarity.)
Regarding Claim 8
Luong in view of Chen discloses: The label generation method according to claim 1, wherein
each of the training data includes image data, and the inference task includes at least one of extracting a region including a feature in the image data and identifying a class of a feature included in the image data. ([Para 0016, 0048-0055], Luong describes classifying image data.),
Regarding Claim 12
Luong in view of Chen discloses: A model generation method in which a computer executes steps of: ([Abstract, Para 0110-0115, and Fig 9], Luong describes a computer for training models and generating pseudo-labels.) acquiring a plurality of third data sets generated by associating the third correct answer label generated by the label generation method according to claim 1 with the third training data; and performing machine learning of a third inference model by using the plurality of acquired third data sets, the machine learning being configured by training, for each of the third data sets, the third inference model such that an inference result obtained by performing the inference task by the third inference model on the third training data conforms to a correct answer indicated by the third correct answer label. ([Para 0009-0013, and Para 0058], Luong describe a third machine learning model)
Regarding Claim 15
Luong in view of Chen discloses: A model generation device comprising: a data acquisition unit configured to ([Abstract, Para 0110-0115, and Fig 9], Luong describes a computer for training models and generating pseudo-labels.) acquire a plurality of third data sets generated by associating the third correct answer label generated by the label generation method according to claim 1 with the third training data; and a learning processing unit configured to perform machine learning of a third inference model by using the plurality of acquired third data sets, the machine learning being configured by training, for each of the third data sets, the third inference model such that an inference result obtained by performing the inference task by the third inference model on the third training data conforms to a correct answer indicated by the third correct answer label. ([Para 0009-0013, and Para 0058], Luong describe a third machine learning model)
Regarding Claim 16
Luong in view of Chen discloses: A non-transitory computer readable medium storing a model generation program, the model generation program configured to cause a computer to execute steps of: ([Abstract, Para 0110-0115, and Fig 9], Luong describes a computer for training models and generating pseudo-labels.) acquiring a plurality of third data sets generated by associating the third correct answer label generated by the label generation method according to claim 1 with the third training data; and performing machine learning of a third inference model by using the plurality of acquired third data sets, the machine learning being configured by training, for each of the third data sets, the third inference model such that an inference result obtained by performing the inference task by the third inference model on the third training data conforms to a correct answer indicated by the third correct answer label. ([Para 0009-0013, and Para 0058], Luong describe a third machine learning model)
Claim(s) 2, and 10-11 is/are rejected under 35 U.S.C. 103 as being unpatentable over Luong in view of Chen and Meng et al. (US 20200334538 A1, hereinafter "Meng").
Regarding Claim 2
Luong in view of Chen discloses: The label generation method according to claim 1,
Luong in view of Chen does not explicitly disclose: wherein the third training data is acquired in a target domain different from the source domain.
However, Meng discloses in the same field of endeavor: wherein the third training data is acquired in a target domain different from the source domain. ([Para 0033-0037 and Fig 2] describes target domain data different from the source domain data.)
It would have been obvious to a person of ordinary skill in art before the effective filling date of the invention to implement the function of Speech recognition disclosed by Meng into the method of Luong in view of Chen to identify class in sound data. The modification would have been obvious because one of the ordinary skills of the art would be motivated to utilize the feature of Speech recognition disclosed by Meng as all the references are in the field of machine learning. A person of ordinary skill of the art would have been motivated to perform the combination for being able to receive different domains data that are fed in parallel to a teacher model and student model respectively.
Regarding Claim 10
Luong in view of Chen discloses: The label generation method according to claim 1, wherein
Luong in view of Chen does not explicitly disclose: each of the training data includes sound data, and the inference task includes at least one of extracting a region including a feature in the sound data and identifying a class of a feature included in the sound data.
However, Meng discloses in the same field of endeavor: each of the training data includes sound data, and the inference task includes at least one of extracting a region including a feature in the sound data and identifying a class of a feature included in the sound data. ([Para 0031-0033, 0041, 0050, 0060] describe speech recognition models trained on sounds data.)
It would have been obvious to a person of ordinary skill in art before the effective filling date of the invention to implement the function of Speech recognition disclosed by Meng into the method of Luong in view of Chen to identify class in sound data. The modification would have been obvious because one of the ordinary skills of the art would be motivated to utilize the feature of Speech recognition disclosed by Meng as all the references are in the field of machine learning. A person of ordinary skill of the art would have been motivated to perform the combination for being able to classify sound data.
Regarding Claim 11
Luong in view of Chen and Meng discloses: The label generation method according to claim 1, wherein each of the training data includes sensing data, and the inference task includes at least one of extracting a region including a feature in the sensing data and identifying a class of a feature included in the sensing data. ([Para 0031-0033, 0041, 0050, 0060], Meng describe models trained on sounds data (i.e. sensing data).)
Claim(s) 3 is/are rejected under 35 U.S.C. 103 as being unpatentable over Luong in view of Chen, Meng, and Ojha et al. (US 20220254071 A1, hereinafter "Ojha").
Regarding Claim 3
Luong in view of Chen and Meng discloses: The label generation method according to claim 2, wherein the applying a disturbance to the first training data is configured by transforming the first training data using a trained transformation model, ([Para 0037-0039 and Fig 1-3], Luong describes a trained second machine learning model injecting noise to the dataset.)
Luong in view of Chen and Meng does not explicitly disclose: the trained transformation model is generated to acquire an ability to transform a style of the first training data into a style of the third training data by machine learning using the first training data and the third training data.
However, Ojha discloses in the same field of endeavor: the applying a disturbance to the first training data is configured by transforming the first training data using a trained transformation model, and the trained transformation model is generated to acquire an ability to transform a style of the first training data into a style of the third training data by machine learning using the first training data and the third training data. ([Para 0049-0051, 0077, 0081-0082, 0108] describes GAN translation system modifies a generative adversarial neural network to generate digital images having a style or appearance of a target domain and match styles.)
It would have been obvious to a person of ordinary skill in art before the effective filling date of the invention to implement the function of GAN-to-GAN Translation disclosed by Ojha into the method of Luong in view of Chen and Meng to transform a style of data. The modification would have been obvious because one of the ordinary skills of the art would be motivated to utilize the feature of GAN-to-GAN Translation disclosed by Ojha as all the references are in the field of machine learning. A person of ordinary skill of the art would have been motivated to perform the combination for being able to generate target neural networks that have cross-domain correspondences with a source neural network.
Claim(s) 4 is/are rejected under 35 U.S.C. 103 as being unpatentable over Luong in view of Chen, Meng, and Al-Turki et al. (US 10,839,269 B1, hereinafter "Al-Turki").
Regarding Claim 4
Luong in view of Chen and Meng discloses: The label generation method according to claim 2, wherein
Luong in view of Chen and Meng does not explicitly disclose: the first inference model and the second inference model are further trained by adversarial learning with an identification model, and the adversarial learning includes: training the identification model using the first training data and the third training data to identify which training data of the first training data and the third training data an inference result of the first inference model is for; training the first inference model using the first training data and the third training data to degrade identification performance of the identification model; training the identification model using the second training data and the third training data to identify which training data of the second training data and the third training data an inference result of the second inference model is for; and training the second inference model using the second training data and the third training data to degrade identification performance of the identification model.
However, Al-Turki discloses in the same field of endeavor: the first inference model and the second inference model are further trained by adversarial learning with an identification model ([Col 5 lines 60-67 and Fig 1-3] describes adversarial learning between a generator and domain discriminator.), and
the adversarial learning includes:
training the identification model using the first training data and the third training data to identify which training data of the first training data and the third training data an inference result of the first inference model is for; ([Col 7 lines 20-67, claims 1-8, Algorithm 1, and Fig 1-3] describes the domain discriminator receives source features and target features to identify a domain.),
training the first inference model using the first training data and the third training data to degrade identification performance of the identification model; ([Col 7 lines 20-67, claims 1-8, Algorithm 1, and Fig 1-3] describes a generator trained adversely to fool the discriminator (i.e. degrading identification).)
training the identification model using the second training data and the third training data to identify which training data of the second training data and the third training data an inference result of the second inference model is for; ([Col 7 lines 20-67, claims 1-8, Algorithm 1, and Fig 1-3] describes the domain discriminator receiving features and identifying a domain it belongs to.), and
training the second inference model using the second training data and the third training data to degrade identification performance of the identification model. ([Col 7 lines 20-67, claims 1-8, Algorithm 1, and Fig 1-3] describes target napping is adversely trained against the discriminator.)
It would have been obvious to a person of ordinary skill in art before the effective filling date of the invention to implement the function of Generative Adversarial Networks disclosed by Al-Turki into the method of Luong in view of Chen and Meng to further train by adversarial learning with an identification model. The modification would have been obvious because one of the ordinary skills of the art would be motivated to utilize the feature of Generative Adversarial Networks disclosed by Al-Turki as all the references are in the field of machine learning. A person of ordinary skill of the art would have been motivated to perform the combination for being able to decrease cross - domain distribution difference and ensure high accuracy and stability of source classifier and thus achieves better classification performance on target data.
Claim(s) 9 is/are rejected under 35 U.S.C. 103 as being unpatentable over Luong in view of Chen, Meng, and Wang et al. ("Weakly Supervised Adversarial Domain Adaptation for Semantic Segmentation in Urban Scenes", hereinafter "Wang").
Regarding Claim 9
Luong in view of Chen and Meng discloses: The label generation method according to claim 2, wherein each of the training data includes image data, ([Para 0016, 0048-0055], Luong describes classifying image data.)
Luong in view of Chen and Meng does not explicitly disclose: the inference task includes extracting a region including a feature in the image data, the first inference model and the second inference model are further trained by adversarial learning with an identification model, and the adversarial learning includes: training the identification model using the first training data and the third training data to identify for each pixel which training data of the first training data and the third training data an inference result of the first inference model is for; training the first inference model using the first training data and the third training data to degrade identification performance of the identification model; training the identification model using the second training data and the third training data to identify for each pixel which training data of the second training data and the third training data an inference result of the second inference model is for; and training the second inference model using the second training data and the third training data to degrade identification performance of the identification model.
However, Wang discloses in the same field of endeavor:
the inference task includes extracting a region including a feature in the image data, ([Section III and Fig 2] describes a detection and segmentation model (DS model) that perform object detection and segmenting.)
the first inference model and the second inference model are further trained by adversarial learning with an identification model, ([Section III and Fig 2] describes segmentation model is adversarial trained against the domain discriminator (i.e. identification model).) and
the adversarial learning includes:
training the identification model using the first training data and the third training data to identify for each pixel which training data of the first training data and the third training data an inference result of the first inference model is for; ([Section III and Fig 2] describes a pixel-level domain classifier identifies source domain form target domain images features.)
training the first inference model using the first training data and the third training data to degrade identification performance of the identification model; ([Section III and Fig 2] describes the DS model as a generator adversarialy trained to learn domain-invariant features.)
training the identification model using the second training data and the third training data to identify for each pixel which training data of the second training data and the third training data an inference result of the second inference model is for; ([Section III and Fig 2] describes a pixel-level domain classifier identifies source domain form target domain images features.) and
training the second inference model using the second training data and the third training data to degrade identification performance of the identification model. ([Section III and Fig 2] describes the DS model as a generator adversarialy trained to learn domain-invariant features.)
It would have been obvious to a person of ordinary skill in art before the effective filling date of the invention to implement the function of a pixel-level domain classifier disclosed by Wang into the method of Luong in view of Chen and Meng to further train by adversarial learning with an identification model. The modification would have been obvious because one of the ordinary skills of the art would be motivated to utilize the feature of pixel-level domain classifier disclosed by Wang as all the references are in the field of machine learning. A person of ordinary skill of the art would have been motivated to perform the combination for being able to distinguish image features from different domains.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure. Xu et al. (US 20210390355 A1) describes source and target models. Chidlovski (US 20210019629 A1) describes source and target image classifiers.
THIS ACTION IS MADE FINAL. Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any nonprovisional extension fee (37 CFR 1.17(a)) pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the mailing date of this final action.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to TEWODROS E MENGISTU whose telephone number is (571)270-7714. The examiner can normally be reached Mon-Fri 9:30-5:30.
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, ABDULLAH KAWSAR can be reached at (571)270-3169. 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.
/TEWODROS E MENGISTU/Primary Examiner, Art Unit 2127
Prosecution Insights