Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Response to Amendment
The Amendment filed 05/19/2026 has been entered. Claims 1-20 remain pending in the application.
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 1-7, 12 and 15 are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more. The analysis of the claims will follow the 2019 Revised Patent Subject Matter Eligibility Guidance, 84 Fed. Reg. 50 (“2019 PEG”)
Claims 1, 12 and 15 have the following abstract idea analysis.
Step 1: The claim is directed to “a method and apparatus”. The claims are directed to the statutory categories accordingly.
Step 2A Prong 1: claims recite the abstract idea limitations of "calculate a geometric center of the embedding vector ", " convert the embedding vector obtained by the first processing layer into a vector expressed", "calculating the gradient" and "converting the gradient to a gradient expressed in the hyperbolic space". These limitations include mathematical concepts see MPEP § 2106.04(a)(2)) where it cites "the phrase “calculating the force of the object by multiplying its mass by its acceleration” is using a textual replacement for the particular equation " and "a conversion between binary coded decimal and pure binary". The specification also provides example calculation of a midpoint and conversion formula (See USPGPUB ¶210 and ¶215). Thus, these steps are an abstract idea in the “mathematical concept”. Other sections of the claims such as "obtaining to-be-processed data", processing the to-be-processed data", "outputting the processing result", "extracting", "classifying", "the neural network comprises a feature extraction network and a classification network", "a conformal conversion layer" and "updating the neural network" are advanced processes, too generic or high level to be listed as a judicial exception given the available descriptions and MPEP comparisons.
Step 2A Prong 2: The judicial exceptions recited in these claims are not integrated into a practical application. Merely invoking "a trained neural network", "to be processed data", "a processor", or "memory" does not yield eligibility. Claims are still in line with mental concepts such as claim 1-7, 12 and 15 are not specific to a practical application. The additional elements as such are processors and instructions which do not include specialized hardware. See MPEP § 2106.05(f).
Claim 1-7, 12 and 15 do not include a particular field but even doing so may not be sufficient to overcome the abstract idea rejection. Merely applying an model to a field or data without an advancement in the new field or new hardware is ineligible. MPEP § 2106.05(h).
Step 2B: The claims do not contain significantly more than their judicial exceptions. Processors, memory and other hardware are in their standard forms in the field. These additional elements are well-understood, routine, and conventional activity, see MPEP 2106.05(d)(II). Claims lacks any particular "how" or algorithm for a solution in a field in a novel way. Claims require more specificity on processes that would be incapable of simple mathematics, mental processes or use more substantial structure than conventional devices such as non-textbook implementations.
Regarding claims 3 and 5-7, they merely narrow the previously recited abstract idea limitations with more abstract concepts and/or routine fundamental processes. For the reasons described above with respect to claim 1 this judicial exception is not meaningfully integrated into a practical application, or significantly more than the abstract idea. Abstract idea steps 1, 2A prong 1 and 2 remain the same as independent analysis above. See specification for more practical application concepts as none are seen in claims 3 and 5-7.
With respect to step 2B These claims disclose similar limitations described for the dependent claims above and do not provide anything significantly more than mathematical or mental concepts. Claims 2-30, 32-35 and 37-41 recite the additional elements of "wherein the classification network comprises a plurality of neurons, each neuron is configured to process input data based on an activation function, and the activation function comprises the operation rule based on the hyperbolic space. wherein the embedding vector is expressed based on a first conformal model; the feature extraction network further comprises a conformal conversion layer; the conformal conversion layer is configured to convert the embedding vector obtained by the first processing layer into a vector expressed based on a second conformal model, and input the vector expressed based on the second conformal model to the second processing layer; the second processing layer is configured to calculate a geometric center of the vector expressed based on the second conformal model, to obtain the feature vector; and the conformal conversion layer is further configured to convert the feature vector obtained by the second processing layer into a vector expressed based on the first conformal model, and input the vector expressed based on the first conformal model to the classification network, wherein the first conformal model represents that the hyperbolic space is mapped to Euclidean space in a first conformal mapping manner, and the second conformal model represents that the hyperbolic space is mapped to the Euclidean space in a second conformal mapping manner. wherein the embedding vector is expressed based on a second conformal model; the second processing layer is configured to calculate a geometric center of the embedding vector expressed based on the second conformal model, to obtain the feature vector, wherein the second conformal model represents that the hyperbolic space is mapped to Euclidean space in a second conformal mapping manner. wherein the classification network is configured to: process the feature vector based on the operation rule of the hyperbolic space to obtain a to-be-normalized vector expressed in the hyperbolic space; and map the to-be-normalized vector to the Euclidean space, and perform normalization processing on the to-be-normalized vector mapped to the Euclidean space, to obtain the processing result." These elements are more abstract concepts, generic applications to a field of use or well-understood, routine, conventional activity (see MPEP § 2106.05(d) and can't be simply appended to qualify as significantly more or being a practical application. What type of application, or structure of components beyond generic machine learning is still unknown for these claims. Therefore claims 3 and 5-7 also recites abstract ideas that do not integrate into a practical application or amount to significantly more than the judicial exception, and are rejected under U.S.C. 101.
Claim Rejections - 35 USC § 103
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
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 of this title, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
Claims 8-12 and 16-18 are rejected under 35 U.S.C. 103 as being unpatentable over Sikka et al. (US 20190325342 A1 hereinafter Sikka) in view of Gao et al. (US 20170032035 A1 hereinafter Gao).
As to independent claim 8, Sikka teaches a data processing method, wherein the method comprises: [process data via embedding ¶5-6]
obtaining training data and a corresponding category label; [images and tags/labels ¶41 " For training, a database with images having semantic tags may be used. Keywords from image captions may then be extracted and used as labels for training the images."]
processing the training data by using a neural network, to obtain a processing result, wherein [training for results Fig. 7 ¶41 "For training, a database with images having semantic tags may be used. Keywords from image captions may then be extracted and used as labels for training the images. A ranking loss algorithm can be adjusted to push similar images and tags (words) together and vice-versa. The mean average precision (MAP) may be output for evaluating the training results."]
the feature extraction network is configured to extract a feature vector of the training data, and [embeddings are part the network that extracts vector into non-Euclidean (hyperbolic ¶7) space ¶37 "In block 310, the first modality feature vector of the multimodal content and the second modality feature vector of the multimodal content are semantically embedded in a non-Euclidean geometric space"] the classification network is configured to process the feature vector based on an operation rule of hyperbolic space, to obtain the processing result; [part of network is for categorizing (classifies) input image features into a result (i.e. "land animals") ¶33 "using a non-Euclidean space such as, for example, a Poincaré space allows distinct classes to form in broader categories such as “plants” and “land animals.""], [uses warping (rule) ¶34 "hyperbolic embeddings provide a way to capture distances that grow exponentially through a logarithm-like warping of distance space."]
obtaining a loss based on the category label and the processing result; [loss and tags ¶41 "
A ranking loss algorithm can be adjusted to push similar images and tags (words) together and vice-versa."]
obtaining, based on the loss, a gradient expressed in the hyperbolic space; and [loss with gradient decent in Riemannian (hyperbolic) ¶40-41 " A contrastive loss function with Riemannian SGD (Stochastic Gradient Descent) was used for the embedding"]
updating the neural network based on the gradient to obtain an updated neural network. [continue training (update) for results Fig. 7 ¶41 "For training, a database with images having semantic tags may be used. Keywords from image captions may then be extracted and used as labels for training the images. A ranking loss algorithm can be adjusted to push similar images and tags (words) together and vice-versa. The mean average precision (MAP) may be output for evaluating the training results."]
Sikka does not specifically teach the neural network comprises a feature extraction network and a classification network.
However, Gao teaches the neural network comprises a feature extraction network and a classification network, [separate sections for classification (DNN) and feature extractions (Fig.6 602), ¶19-20 " a deep structured semantic model (DSSM) may be used to project an input item to an output item in a semantic space. For example, the input item may correspond to an input vector that represents one or more words, while the output item may correspond to a concept vector that expresses semantic information regarding the word(s)"…" a DSSM comprises a pair of DNNs, where one DNN may be used for mapping the source (e.g., text) into a semantic vector"]
Accordingly, it would have been obvious to a person of ordinary skill in the art before the effective filling date of the claimed invention to modify the models disclosed by Sikka by incorporating the neural network comprises a feature extraction network and a classification network disclosed by Gao because both techniques address the same field of data machine learning and by incorporating Gao into Sikka provide more relevant results with faster convergence [Gao ¶78]
As to dependent claim 9, the rejection of claim 8 is incorporated, Sikka and Gao further teach wherein the updating the neural network based on the gradient to obtain an updated neural network comprises: updating the feature extraction network in the neural network based on the gradient, to obtain an updated feature extraction network, wherein the updated feature extraction network is configured to extract the feature vector expressed by the training data in the hyperbolic space. [Sikka hyperbolic embedding over iterations (updating) Fig. 7 "Iterations" and SGD ¶40-¶41 "Poincaré ball is a realization of hyperbolic space (open d dimensional unit ball) and, in an embodiment, the Poincaré ball can be used to model the hyperbolic embedding space"]
As to dependent claim 10, the rejection of claim 8 is incorporated, Sikka and Gao further teach wherein the classification network is configured to: process the feature vector based on the operation rule of the hyperbolic space to obtain a to-be-normalized vector expressed in the hyperbolic space; and [Sikka vectors ready for normalizing and categories for classification ¶56-57, ¶40 "Hierarchies are determined by the normalization of the embedded vectors. As illustrated in a view 400A of FIG. 4A, Euclidean space does not inherently preserve hierarchies as the content is spread across a single plane"]
map the to-be-normalized vector to the Euclidean space, and perform normalization processing on the to-be-normalized vector mapped to the Euclidean space, to obtain the processing result. [Sikka normalizing and Fig. 4A 400A illustrates mapping to Euclidean ¶56-57, ¶40 "400A of FIG. 4A, Euclidean space does not inherently preserve hierarchies as the content is spread across a single plane"]
As to dependent claim 11, the rejection of claim 10 is incorporated, Sikka and Gao further teach wherein the obtaining a loss based on the category label and the processing result comprises: obtaining the loss based on the category label, the processing result, and a target loss function, wherein the target loss function is a function expressed in the Euclidean space. [Sikka loss and tags with standard loss ¶41, 45 " A ranking loss algorithm can be adjusted to push similar images and tags (words) together and vice-versa."]
As to dependent claim 12, the rejection of claim 10 is incorporated, Sikka and Gao further teach calculating the gradient corresponding to the loss, wherein the gradient is expressed in the Euclidean space; [Gao cross-entropy SGD loss ¶76-77], [Sikka ¶36 "word2vec (Euclidean space) may be used to provide the first feature vector for text modalities. However, the inventors have found that performance may be increased by retraining the word2vec with vectors from a non-Euclidean space."]
converting the gradient to a gradient expressed in the hyperbolic space; and [Sikka alter gradient for the manifold (hyperbolic) from word2vec (Euclidean) ¶36, ¶41 " The structure of loss function and the gradient descent are altered to create a linear projection layer to constrain embedding vectors to the manifold. In one example, a pre-trained word2vec model may be used and the results can be projected to the manifold via a few non-linear layers. "]
updating the neural network based on the gradient expressed in the hyperbolic space. [Sikka continue training for results Fig. 7 ¶41 "For training, a database with images having semantic tags may be used. Keywords from image captions may then be extracted and used as labels for training the images. A ranking loss algorithm can be adjusted to push similar images and tags (words) together and vice-versa. The mean average precision (MAP) may be output for evaluating the training results."]
As to independent claim 16, Sikka teaches a data processing apparatus, wherein the apparatus comprises a memory and a processor, the memory stores code, and the processor is configured to execute the code to perform: [apparatus, processor and memory with instructions ¶8]
obtaining training data and a corresponding category label; [images and tags/labels ¶41 " For training, a database with images having semantic tags may be used. Keywords from image captions may then be extracted and used as labels for training the images."]
processing the training data by using a neural network, to obtain a processing result, wherein [training for results Fig. 7 ¶41 "For training, a database with images having semantic tags may be used. Keywords from image captions may then be extracted and used as labels for training the images. A ranking loss algorithm can be adjusted to push similar images and tags (words) together and vice-versa. The mean average precision (MAP) may be output for evaluating the training results."]
the feature extraction network is configured to extract a feature vector of the training data, and [embeddings are part the network that extracts vector into non-Euclidean (hyperbolic ¶7) space ¶37 "In block 310, the first modality feature vector of the multimodal content and the second modality feature vector of the multimodal content are semantically embedded in a non-Euclidean geometric space"] the classification network is configured to process the feature vector based on an operation rule of hyperbolic space, to obtain the processing result; [part of network is for categorizing (classifies) input image features into a result (i.e. "land animals") ¶33 "using a non-Euclidean space such as, for example, a Poincaré space allows distinct classes to form in broader categories such as “plants” and “land animals.""], [uses warping (rule) ¶34 "hyperbolic embeddings provide a way to capture distances that grow exponentially through a logarithm-like warping of distance space."]
obtaining a loss based on the category label and the processing result; [loss and tags ¶41 "
A ranking loss algorithm can be adjusted to push similar images and tags (words) together and vice-versa."]
obtaining, based on the loss, a gradient expressed in the hyperbolic space; and [loss with gradient decent in Riemannian (hyperbolic) ¶40-41 " A contrastive loss function with Riemannian SGD (Stochastic Gradient Descent) was used for the embedding"]
updating the neural network based on the gradient to obtain an updated neural network. [continue training (update) for results Fig. 7 ¶41 "For training, a database with images having semantic tags may be used. Keywords from image captions may then be extracted and used as labels for training the images. A ranking loss algorithm can be adjusted to push similar images and tags (words) together and vice-versa. The mean average precision (MAP) may be output for evaluating the training results."]
Sikka does not specifically teach the neural network comprises a feature extraction network and a classification network.
However, Gao teaches the neural network comprises a feature extraction network and a classification network, [separate sections for classification (DNN) and feature extractions (Fig.6 602), ¶19-20 " a deep structured semantic model (DSSM) may be used to project an input item to an output item in a semantic space. For example, the input item may correspond to an input vector that represents one or more words, while the output item may correspond to a concept vector that expresses semantic information regarding the word(s)"…" a DSSM comprises a pair of DNNs, where one DNN may be used for mapping the source (e.g., text) into a semantic vector"]
Accordingly, it would have been obvious to a person of ordinary skill in the art before the effective filling date of the claimed invention to modify the models disclosed by Sikka by incorporating the neural network comprises a feature extraction network and a classification network disclosed by Gao because both techniques address the same field of data machine learning and by incorporating Gao into Sikka provide more relevant results with faster convergence [Gao ¶78]
As to dependent claim 17, the rejection of claim 16 is incorporated, Sikka and Gao further teach wherein the processor is configured to obtain the code and perform: updating the feature extraction network in the neural network based on the gradient, to obtain an updated feature extraction network, wherein the updated feature extraction network is configured to extract the feature vector expressed by the training data in the hyperbolic space. [Sikka hyperbolic embedding over iterations (updating) Fig. 7 "Iterations" and SGD ¶40-¶41 "Poincaré ball is a realization of hyperbolic space (open d dimensional unit ball) and, in an embodiment, the Poincaré ball can be used to model the hyperbolic embedding space"]
As to dependent claim 18, the rejection of claim 16 is incorporated, Sikka and Gao further teach wherein the classification network is configured to: process the feature vector based on the operation rule of the hyperbolic space to obtain a to-be-normalized vector expressed in the hyperbolic space; and [Sikka vectors ready for normalizing and categories for classification ¶56-57, ¶40 "Hierarchies are determined by the normalization of the embedded vectors. As illustrated in a view 400A of FIG. 4A, Euclidean space does not inherently preserve hierarchies as the content is spread across a single plane"]
map the to-be-normalized vector to the Euclidean space, and perform normalization processing on the to-be-normalized vector mapped to the Euclidean space, to obtain the processing result. [Sikka normalizing and Fig. 4A 400A illustrates mapping to Euclidean ¶56-57, ¶40 "400A of FIG. 4A, Euclidean space does not inherently preserve hierarchies as the content is spread across a single plane"]
Response to Arguments
Applicant's arguments filed 06/16/2026. In the remark, with respect to 101, applicant argues that:
As discussed in MPEP 2106.04(d)(III), the claims are clearly directed toward "improvements as to how the machine learning model itself operates", which is comparable to improving the functioning of a computer. This is also clear from the Specification. Thus, the claims integrate the alleged judicial exception into a practical application, and the rejection should be withdrawn.
As to point (1), Applicant’s arguments with respect to claim 1, Examiner respectfully disagrees with Applicant's arguments.
Examiner still believes the claims are directed to an abstract idea without significantly more. The amendment imports limitations directed to obtaining an embedding vector and calculating a geometric center. These define mathematical processing but do not apply that processing in a concrete technological process. Adding more detail to the mathematical calculation does not, by itself, integrated the exception into a practical application. The claim still broadly recites obtaining data, mathematical processing, and outputting results. These elements are still claimed at a high level. Claim 1 does not recite training of the model or otherwise changes how the model operates to solve a specific problem. See MPEP 2106.04(d)(1) and 2106.05(a) including guidance in Ex parte Desjardins. The claims are also unlike USPTO provide example 39 because that claim did not recite mathematical relationships or formulas. Hence, the claims are directed to an abstract idea without significantly more.
Applicant's arguments filed 06/16/2026. In the remark, applicant argues that:
Sikka and Gao fail to teach “the second processing laver is configured to calculate a geometric center of the embedding vector in the hyperbolic space, to obtain the feature vector.” As recited by amended claim 1.
Sikka and Gao fail to teach “obtaining, based on the loss, a gradient expressed in the hyperbolic space; and updating the neural network based on the gradient to obtain an updated neural network.” as recited by original claim 8
As to point (1), Applicant’s arguments with respect to claim 1 have been considered and the 103 rejection is withdrawn.
As to point (2), Sikka and Gao do teach “obtaining, based on the loss, a gradient expressed in the hyperbolic space; and updating the neural network based on the gradient to obtain an updated neural network.” as recited by claim 8 and similarly claim 16. According to MPEP 2111, examiner is obliged to give the terms or phrases their broadest interpretation definition awarded by one of an ordinary skill in the art unless applicant has provided some indication of the definition of the claimed terms or phrases. Claim 8 does not require the claim 12 sequence of calculating the Euclidean gradient, converting and updating. The claim also does not require that the entire neural network be updated in every layer. Updating feature extraction is mentioned as an example in the specification (See PGPUB ¶219-221). Gao also teaches supervised training with queries and labels including cross-entropy loss for query classification (see ¶17, ¶77). Sikka also expressly states that a loss function with Riemannian SGD (gradients) is used for embedding. (see ¶40). This is part of Pointcare/hyperbolic embedding in non-Euclidean space (See ¶33). In response to Applicants’ arguments against the references individually, one cannot show non obviousness by attacking references individually where the rejections are based on combinations of references. Hence, Sikka and Gao do teach “obtaining, based on the loss, a gradient expressed in the hyperbolic space; and updating the neural network based on the gradient to obtain an updated neural network.” as recited by claim 8 and similarly claim 16.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure. Applicant is required under 37 C.F.R. § 1.111(c) to consider these references fully when responding to this action.
CREED et al. (US 20210081717 A1) teaches loss-based updating of graph neural networks utilizing embedding (See ¶8).
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 BEAU SPRATT whose telephone number is (571)272-9919. The examiner can normally be reached M-F 8:30-5 EST.
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, Jennifer To can be reached at 5712127212. 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.
/BEAU D SPRATT/Primary Examiner, Art Unit 2143
Prosecution Insights