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 .
Specification
The lengthy specification has not been checked to the extent necessary to determine the presence of all possible minor errors. Applicant’s cooperation is requested in correcting any errors of which applicant may become aware in the specification.
Drawings
The applicant’s submitted drawings appear to be acceptable for examination purposes. Applicant’s cooperation is requested in correcting any errors of which applicant may become aware in the drawings.
Claim Objections
Claim 10 is objected to because of the following informalities: “native Bayes” appears as though it should be “naïve Bayes.” Appropriate correction is required.
Claim 18 is objected to because of the following informalities: “native Bayes” appears as though it should be “naïve Bayes.” Appropriate correction is required.
Claim Rejections - 35 USC § 112
The following is a quotation of 35 U.S.C. 112(b):
(b) CONCLUSION.—The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the inventor or a joint inventor regards as the invention.
The following is a quotation of 35 U.S.C. 112 (pre-AIA ), second paragraph:
The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the applicant regards as his invention.
Claims 5, 14, 16, and 20 are rejected under 35 U.S.C. 112(b) or 35 U.S.C. 112 (pre-AIA ), second paragraph, as being indefinite for failing to particularly point out and distinctly claim the subject matter which the inventor or a joint inventor (or for applications subject to pre-AIA 35 U.S.C. 112, the applicant), regards as the invention.
Claim 5 recites the limitation "wherein freezing index weights in the embedding layer during training of the artificial neural network comprises” in lines 1-2. There is insufficient antecedent basis for this limitation in the claim, as this is the first recitation of freezing index weights.
Claim 14 recites the limitation "the embedding layer” in line 1. There is insufficient antecedent basis for this limitation in the claim (Examiner’s Note: it appears that claim 14 is intended to depend from claim 13, which recites an embedding layer).
Claim 16 depends upon claim 14, and thus includes the aforementioned limitation(s).
Claim 16 also recites the limitation "wherein freezing index weights in the embedding layer during training of the artificial neural network comprises” in lines 1-2. There is insufficient antecedent basis for this limitation in the claim, as this is the first recitation of freezing index weights.
Claim 20 recites the limitation "the set of training vectors” in line 3. There is insufficient antecedent basis for this limitation in the claim.
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 (i.e., changing from AIA to pre-AIA ) 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.
This application currently names joint inventors. In considering patentability of the claims the examiner presumes that the subject matter of the various claims was commonly owned as of the effective filing date of the claimed invention(s) absent any evidence to the contrary. Applicant is advised of the obligation under 37 CFR 1.56 to point out the inventor and effective filing dates of each claim that was not commonly owned as of the effective filing date of the later invention in order for the examiner to consider the applicability of 35 U.S.C. 102(b)(2)(C) for any potential 35 U.S.C. 102(a)(2) prior art against the later invention.
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
The factual inquiries for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
1. Determining the scope and contents of the prior art.
2. Ascertaining the differences between the prior art and the claims at issue.
3. Resolving the level of ordinary skill in the pertinent art.
4. Considering objective evidence present in the application indicating obviousness or nonobviousness.
Claim(s) 1, 2, 6, 11-13, and 17-20 is/are rejected under 35 U.S.C. 103 as being unpatentable over Kung et al. (Efficient Multi-Task Auxiliary Learning: Selecting Auxiliary Data by Feature Similarity, Nov 2021, pgs. 416-428) in view of Jaiswal (US 11,775,617).
As per claim 1, Kung teaches a computer-implemented method comprising: selecting, for training of an artificial neural network, a training batch of points from within a dataset of training points [a multi-task auxiliary learning system that selects training samples most relevant to the primary task (pg. 416, abstract; pg. 420, fig. 2; etc.); where selected training samples are a batch of points from the dataset of training points], and training the artificial neural network using the selected training points [training samples are sent to a task-discriminative network to determine similarity ranking of samples for auxiliary tasks to the primary task, and the most similar (ranked) samples are used to train the (primary) task neural network model (pg. 416, abstract; pgs. 419-420, sections 3.2-3.3 and fig. 2; etc.)].
While Kung teaches selecting batches of training samples for training a primary task neural network, based upon a similarity to the primary task (see above), it has not been relied upon for teaching each training point comprising a plurality of sets of values, where each value corresponds to an index into an embedding space included in the artificial neural network; forming, from the dataset of training points, a neighborhood of training points associated with the training batch such that each member of the neighborhood shares at least one index with at least one training point included in the training batch; choosing, via a cluster analysis method, a cluster of points from the neighborhood of training points associated with the training batch; and training the artificial neural network using the chosen cluster of points from the neighborhood of points associated with the training batch.
Jaiswal teaches each training point comprising a plurality of sets of values, where each value corresponds to an index into an embedding space included in the artificial neural network [an embedding vector representation in the embedding space is created for each sample, which is indexed (col. 14, lines 4-33; fig. 3; etc.)]; forming, from the dataset of training points, a neighborhood of training points associated with the training batch such that each member of the neighborhood shares at least one index with at least one training point included in the training batch [a search index (neighborhood) is generated to find a selection of similar embedding vectors in the embedding space (col. 14, lines 4-33; fig. 3; etc.); where the embedding vectors share the search index in the embedding space; for determining a similarity selection of training samples in the system of Kung, above]; choosing, via a cluster analysis method, a cluster of points from the neighborhood of training points associated with the training batch [the search index (neighborhood) can be used to find embedding vectors in the embedding space with the largest cosine similarity/smallest distance/other similarity score (cluster analysis), as most similar to the embedding being searched (col. 14, lines 4-33; fig. 3; etc.); for determining a similarity selection of training samples in the system of Kung, above]; and training the artificial neural network using the chosen cluster of points from the neighborhood of points associated with the training batch [the search index (neighborhood) can be used to select embedding vectors in the embedding space most similar to the embedding being searched (col. 14, lines 4-33; fig. 3; etc.); for determining a similarity selection of training samples used to train the neural network in the system of Kung, above].
Kung and Jaiswal are analogous art, as they are within the same field of endeavor, namely improving machine learning models and making similarity determinations for data samples, the machine learning models including convolutional neural networks used for object detection/classification/etc.
It would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to use the embedding space similarity search for finding similar data samples, as taught by Jaiswal, for the similarity sampling for selecting training samples to train the neural network in the system taught by Kung.
Jaiswal provides motivation as [using the embedding data allows identification of similar items (col. 14, lines 4-33; fig. 3; etc.) and allows models to handle objects/samples of a wider range of types (col. 1, line 56 to col. 2, line 29; etc.)] where utilizing data from similar tasks/models is also the desired outcome in Kung (see, e.g., pg. 416, abstract) and Kung also provides a suggestion that embeddings may be used to optimize data for faster training (pg. 418, section 2.3; etc.).
As per claim 2, Kung/Jaiswal teaches wherein the artificial neural network comprises an embedding layer [the neural network includes a class-specific embedding layer (Jaiswal: fig. 3; etc.)].
As per claim 6, Kung/Jaiswal teaches choosing, from the set of training points via the cluster analysis method, the cluster of points from the neighborhood of the training batch comprises applying the cluster analysis method within an embedding space of the artificial neural network for each point in the neighborhood of the training batch [the search index (neighborhood) can be used to find embedding vectors in the embedding space with the largest cosine similarity/smallest distance/other similarity score (cluster analysis), as most similar to the search index embedding being searched (Jaiswal: col. 14, lines 4-33; fig. 3; etc.); where searching the shared index includes applying the analysis to each point sharing the index].
As per claim 11, Kung/Jaiswal teaches wherein the cluster analysis method comprises a clustering method based on distance [the search index (neighborhood) can be used to find embedding vectors in the embedding space with the largest cosine similarity/smallest distance/other similarity score (cluster analysis based on distance), as most similar to the search index embedding being searched (Jaiswal: col. 14, lines 4-33; fig. 3; etc.)].
As per claim 12, Kung teaches a system comprising: a selecting module, stored in memory, that selects, for training of an artificial neural network, a training batch of points from within a dataset of training points [using stored program code (a selecting module) (pg. 416, footnote 1; etc.) a multi-task auxiliary learning system selects training samples most relevant to the primary task (pg. 416, abstract; pg. 420, fig. 2; etc.); where selected training samples are a batch of points from the dataset of training points], and a training module, stored in memory, that trains the artificial neural network using the chosen cluster of points from the neighborhood of the training batch [using stored program code (a training module stored in memory) (pg. 416, footnote 1; etc.) training samples are sent to a task-discriminative network to determine similarity ranking of samples for auxiliary tasks to the primary task, and the most similar (ranked) samples are used to train the (primary) task neural network model (pg. 416, abstract; pgs. 419-420, sections 3.2-3.3 and fig. 2; etc.)].
While Kung teaches selecting batches of training samples for training a primary task neural network, based upon a similarity to the primary task (see above), it has not been relied upon for teaching each training point comprising a plurality of sets of values, where each value corresponds to an index into an embedding space included in the artificial neural network; a forming module, stored in memory, that forms, from the dataset of training points, a neighborhood of training points associated with the training batch such that each member of the neighborhood shares at least one index with at least one training point included in the training batch; a choosing module, stored in memory, that chooses, via a cluster analysis method, a cluster of points from the neighborhood of training points associated with the training batch; a training module, stored in memory, that trains the artificial neural network using the chosen cluster of points from the neighborhood of the training batch; and at least one physical processor that executes the selecting module, the forming module, the choosing module, and the training module.
Jaiswal teaches each training point comprising a plurality of sets of values, where each value corresponds to an index into an embedding space included in the artificial neural network [an embedding vector representation in the embedding space is created for each sample, which is indexed (col. 14, lines 4-33; fig. 3; etc.)]; a forming module, stored in memory, that forms, from the dataset of training points, a neighborhood of training points associated with the training batch such that each member of the neighborhood shares at least one index with at least one training point included in the training batch [the models and training may be implemented as instructions stored in memory that are executed by a processor (col. 5, lines 1-17; fig. 1; etc.) where a search index (neighborhood) is generated to find a selection of similar embedding vectors in the embedding space (col. 14, lines 4-33; fig. 3; etc.); where the embedding vectors share the search index in the embedding space; for determining a similarity selection of training samples in the system of Kung, above]; a choosing module, stored in memory, that chooses, via a cluster analysis method, a cluster of points from the neighborhood of training points associated with the training batch [the models and training may be implemented as instructions stored in memory that are executed by a processor (col. 5, lines 1-17; fig. 1; etc.) where the search index (neighborhood) can be used to find embedding vectors in the embedding space with the largest cosine similarity/smallest distance/other similarity score (cluster analysis), as most similar to the embedding being searched (col. 14, lines 4-33; fig. 3; etc.); for determining a similarity selection of training samples in the system of Kung, above]; a training module, stored in memory, that trains the artificial neural network using the chosen cluster of points from the neighborhood of the training batch [the models and training may be implemented as instructions stored in memory that are executed by a processor (col. 5, lines 1-17; fig. 1; etc.) where the search index (neighborhood) can be used to select embedding vectors in the embedding space most similar to the embedding being searched (col. 14, lines 4-33; fig. 3; etc.); for determining a similarity selection of training samples used to train the neural network in the system of Kung, above]; and at least one physical processor that executes the selecting module, the forming module, the choosing module, and the training module [the models and training may be implemented as instructions stored in memory that are executed by a processor (col. 5, lines 1-17; fig. 1; etc.)].
Kung and Jaiswal are analogous art, as they are within the same field of endeavor, namely improving machine learning models and making similarity determinations for data samples, the machine learning models including convolutional neural networks used for object detection/classification/etc.
It would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to use the embedding space similarity search for finding similar data samples, as taught by Jaiswal, for the similarity sampling for selecting training samples to train the neural network in the system taught by Kung.
Jaiswal provides motivation as [using the embedding data allows identification of similar items (col. 14, lines 4-33; fig. 3; etc.) and allows models to handle objects/samples of a wider range of types (col. 1, line 56 to col. 2, line 29; etc.)] where utilizing data from similar tasks/models is also the desired outcome in Kung (see, e.g., pg. 416, abstract) and Kung also provides a suggestion that embeddings may be used to optimize data for faster training (pg. 418, section 2.3; etc.).
As per claim 13, see the rejection of claim 2, above.
As per claim 17, see the rejection of claim 6, above.
As per claim 18, Kung/Jaiswal teaches wherein the cluster analysis method comprises at least one of: a k-nearest neighbor classifier; a nearest centroid classifier; a support vector machine classifier; a native Bayes classifier; or a clustering method based on distance [the search index (neighborhood) can be used to find embedding vectors in the embedding space with the largest cosine similarity/smallest distance/other similarity score (cluster analysis based on distance), as most similar to the search index embedding being searched (Jaiswal: col. 14, lines 4-33; fig. 3; etc.)].
As per claim 19, see the rejection of claim 1, above, wherein Kung/Jaiswal also teaches a non-transitory computer-readable medium comprising computer-readable instructions that, when executed by at least one processor of a computing system, cause the computing system to: [perform the method] [the system can be implemented using stored code/instructions to be executed by a processor (Kung: pg. 416, footnote 1; Jaiswal: col. 5, lines 1-17; fig. 1;etc.)].
As per claim 20, Kung/Jaiswal teaches wherein the computer-readable instructions, when executed by the processor of the computing system, cause the computing system to choose, from the set of training vectors via the cluster analysis method, the cluster of points from the neighborhood of the training batch by applying the cluster analysis method within an embedding space of the artificial neural network for each point in the training batch [the search index (neighborhood) can be used to find embedding vectors in the embedding space with the largest cosine similarity/smallest distance/other similarity score (cluster analysis), as most similar to the search index embedding being searched (Jaiswal: col. 14, lines 4-33; fig. 3; etc.); where searching the shared index includes applying the analysis to each point sharing the index].
Claim(s) 3 and 14 is/are rejected under 35 U.S.C. 103 as being unpatentable over Kung and Jaiswal as applied to claims 2 and 12 above, and further in view of Xian (US 2021/0232908).
As per claim 3, Kung/Jaiswal teaches the computer-implemented method of claim 2, as described above.
While Kung/Jaiswal teaches the neural network including an embedding layer (see above), it has not been relied upon for teaching wherein the embedding layer comprises a matrix comprising a number of rows of index weights corresponding to a number of indices included in the dataset of training points.
Xian teaches wherein the embedding layer comprises a matrix comprising a number of rows of index weights corresponding to a number of indices included in the dataset of training points [the neural network utilizes one or more layers to generate a vector embedding (or latent vector) in the latent space (para. 0037, etc.), using a parameter (weight) matrix of a specified size, where the dimensional size of the parameter matrix can be set to the same dimensional size (number of indices) as the vector embeddings (para. 0082, etc.)].
Kung/Jaiswal and Xian are analogous art, as they are within the same field of endeavor, namely a neural network including embedding layers to generate embedding vectors for input data samples.
It would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to utilize a parameter (weight) matrix for the embedding layer that is the same dimensional size as the vector embeddings, as taught by Xian, for the embedding layer and vector embeddings in the system taught by Kung/Jaiswal.
Xian provides motivation as [using a parameter (weight) matrix allows control of the dimensional size of the parameters and allows maintaining the same dimensional size as the vector embeddings (para. 0082, etc.) which allows more accurate label determinations (abstract, etc.)].
As per claim 14, see the rejection of claim 3, above.
Claim(s) 4 is/are rejected under 35 U.S.C. 103 as being unpatentable over Kung, Jaiswal, and Xian, as applied to claims 3 and 14 above, and further in view of Mithun (US 2021/0089841).
As per claim 4, Kung/Jaiswal/Xian teaches the computer-implemented method of claim 3, as described above.
While Kung/Jaiswal/Xian teaches the neural network including an embedding layer (see above), it has not been relied upon for teaching wherein training the artificial neural network using the chosen cluster of points comprises freezing at least one weight in the embedding layer during training of the artificial neural network.
Mithun teaches wherein training the artificial neural network using the chosen cluster of points comprises freezing at least one weight in the embedding layer during training of the artificial neural network [when training the neural network, the matrices of weights of filters within a portion of the sequence of layers in the network are frozen (paras. 0017, 0156; claim 9; etc.); for the convolutional embedding layer weights in the system of Kung/Jaiswal/Xian, above].
Kung/Jaiswal/Xian and Mithun are analogous art, as they are within the same field of endeavor, namely training neural networks for specific tasks, including convolutional neural networks used for object detection/classification, etc.
It would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to freeze the matrices of weights of filters in a portion of the layers of the network during training, as taught by Mithun, for portions of the layers of the network, including an embedding layer(s), during training in the system taught by Kung/Jaiswal/Xian.
Mithun provides motivation as [freezing weights in specific layers allows training to target/finetune specific portions of the network (para. 0141, 0156, etc.)].
Claim(s) 7-10 is/are rejected under 35 U.S.C. 103 as being unpatentable over Kung and Jaiswal as applied to claim 1 above, and further in view of well-known practices in the art.
As per claim 7, Kung/Jaiswal teaches the computer-implemented method of claim 1, as described above.
While Kung/Jaiswal teaches cluster analysis that can comprise different similarity measures (see above), it has not been relied upon for teaching wherein the cluster analysis method comprises a k-nearest neighbor classifier.
However, the examiner takes official notice that a k-nearest neighbor classifier is old and well-known within the art for determining similarity of data samples for clustering. Therefore, it would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to implement a k-nearest neighbor classifier in the cluster analysis of the system of Kung/Jaiswal, to achieve the predictable result of using a well-known and effective model(s) for accurately determining similarity between data samples in the embedding space of Kung/Jaiswal.
As per claim 8, Kung/Jaiswal teaches the computer-implemented method of claim 1, as described above.
While Kung/Jaiswal teaches cluster analysis that can comprise different similarity measures (see above), it has not been relied upon for teaching wherein the cluster analysis method comprises a nearest centroid classifier.
However, the examiner takes official notice that a nearest centroid classifier is old and well-known within the art for determining similarity of data samples for clustering. Therefore, it would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to implement a nearest centroid classifier in the cluster analysis of the system of Kung/Jaiswal, to achieve the predictable result of using a well-known and effective model(s) for accurately determining similarity between data samples in the embedding space of Kung/Jaiswal.
As per claim 9, Kung/Jaiswal teaches the computer-implemented method of claim 1, as described above.
While Kung/Jaiswal teaches cluster analysis that can comprise different similarity measures (see above), it has not been relied upon for teaching wherein the cluster analysis method comprises a support vector machine classifier.
However, the examiner takes official notice that a support vector machine classifier is old and well-known within the art for determining similarity of data samples for clustering. Therefore, it would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to implement a support vector machine classifier in the cluster analysis of the system of Kung/Jaiswal, to achieve the predictable result of using a well-known and effective model(s) for accurately determining similarity between data samples in the embedding space of Kung/Jaiswal.
As per claim 10, Kung/Jaiswal teaches the computer-implemented method of claim 1, as described above.
While Kung/Jaiswal teaches cluster analysis that can comprise different similarity measures (see above), it has not been relied upon for teaching wherein the cluster analysis method comprises a native Bayes classifier.
However, the examiner takes official notice that a native (or naïve – see above) Bayes classifier is old and well-known within the art for determining similarity of data samples for clustering. Therefore, it would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to implement a native Bayes classifier in the cluster analysis of the system of Kung/Jaiswal, to achieve the predictable result of using a well-known and effective model(s) for accurately determining similarity between data samples in the embedding space of Kung/Jaiswal.
Claim(s) 15 is/are rejected under 35 U.S.C. 103 as being unpatentable over Kung and Jaiswal, as applied to claim 13 above, and further in view of Mithun (US 2021/0089841).
As per claim 15, see the rejection of claim 4, above.
Allowable Subject Matter
Claims 5 and 16 are objected to as being dependent upon a rejected base claim, but would be allowable if rewritten in independent form including all of the limitations of the base claim and any intervening claims. The cited art teaches various systems/methods that include weight matrices in the embedding layer, as well as freezing weights of portions/layers of the neural network during training. However, none of the cited art appears to provide motivation to modify the art cited above to freeze rows included in the embedding layer except rows that correspond to at least one index included in the training batch.
Conclusion
The following is a summary of the treatment and status of all claims in the application as recommended by M.P.E.P. 707.07(i): claims 1-20 are rejected.
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
Bileschi (US 2022/0172055) – discloses a system/method that utilizes a clustering engine to identify a specific number of “neighboring” training sequence embeddings that are closest to a specified sequence embedding.
Li (US 11,200,284) – discloses an embedding space in which embeddings with similar values at the same indices are considered similar data samples.
Anil (US 2024/0078379) – discloses an attention neural network that learns a codebook representation of embeddings and using a distance measure to assign embeddings to a cluster index that maps to a selected code word.
D’Innocente (US 11,809,520) – discloses semantic retrieval including generating a search index to find embedding vectors with a largest similarity/smallest distance, similar to Jaiswal, above.
Zheng (US 2022/0405580) – discloses a neural network that includes mapping user vectors between embeddings by comparing respective embeddings of one row with embeddings of another row to determine similarity of contexts of the rows.
Bharathy (US 2022/0179910) – discloses graph queries to determine results from a neural network, including determining a specific number of the most similar embeddings using embedding indexing.
Xian (US 2021/0264244) – similar to Xian, above, including determining distance values between vector embeddings of a column and embeddings of candidate labels for the column.
Gonzalez et al. (Faster Training by Selecting Samples Using Embeddings, Sept 2018, pgs. 1-7) – discloses improving neural network training by filtering training samples based upon comparing sample embeddings (but appears to use filtering to find more distant/dissimilar samples for training).
Wu et al. (Sampling Matters in Deep Embedding Learning, Jan 2018, pgs. 1-10) – discloses improving training by selecting training samples based on relative distance in the embedding space (similar to Gonzalez, above, it appears to use distance to select more distant/dissimilar samples for training).
Athitsos et al. (BoostMap: An Embedding Method for Efficient Nearest Neighbor Retrieval, Jan 2008, pgs. 89-104) – discloses choosing training triples randomly using specified constraints, including clustering/k-nearest neighbor constraints.
Goutam et al. (LayerOut: Freezing Layers in Deep Neural Networks, Sept 2020, pgs. 1-9) – discloses freezing weights of random layers during training.
Liu et al. (AutoFreeze: Automatically Freezing Model Blocks to Accelerate Fine-tuning, April 2021, pgs. 1-27) – discloses adaptively selecting layers to freeze during training.
The examiner requests, in response to this Office action, that support be shown for language added to any original claims on amendment and any new claims. That is, indicate support for newly added claim language by specifically pointing to page(s) and line number(s) in the specification and/or drawing figure(s). This will assist the examiner in prosecuting the application.
When responding to this office action, Applicant is advised to clearly point out the patentable novelty which he or she thinks the claims present, in view of the state of the art disclosed by the references cited or the objections made. He or she must also show how the amendments avoid such references or objections. See 37 CFR 1.111(c).
Any inquiry concerning this communication or earlier communications from the examiner should be directed to GEORGE GIROUX whose telephone number is (571)272-9769. The examiner can normally be reached M-F 10am-6pm.
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, Omar Fernandez Rivas can be reached at 571-272-2589. 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.
/GEORGE GIROUX/Primary Examiner, Art Unit 2128