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 .
DETAILED ACTION
This is Non-Final Office Action, in responses to Patent Application filed 04/13/2023; is a Continuation in Part of 18299841 , filed on 04/13/2023 PCT/US23/23831 filed on 05/30/2023 is a of 18299841, filed on 04/13/2023. Claim(s) 1-21 are pending. Claim(s) 1, 13 and 21 is/are independent.
In addition, 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.
Information Disclosure Statement
A signed and dated copy of applicant’s IDS, which was filed 01/17/2024, 11/13/2023 and 06/03/2024 is/are attached to this Office Action.
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.
Claim(s) 1-21 fail to recite statutory subject matter, as defined in 35 U.S.C. 101, because: The claimed invention is/are directed to a judicial exception (i.e., abstract idea) without significantly more.
Step 1: YES (Claim(s) is/are process, machine, manufacture or composition of the matter). ... a computer-implemented asymmetric dual encoder... comprising: a token embedder layer section having a first token embedding section associated with a first input and a second token embedding section associated with a second input;
an encoder layer section having a first encoder section configured to receive token embeddings from the first token embedding section and a second encoder section configured to receive token embeddings from the second token embedding section;
a projection layer configured to receive encodings from both the first and second encoder sections and to generate a set of projections, wherein the projection layer is shared by the asymmetric dual encoder system; and
an embedding space configured, based on the set of projections, to generate a question embedding and an answer embedding, the question and answer embeddings for use in identifying a set of candidate answers to an input answer... and therefore, fall into one of the four categories of patent eligible subject matter (process, machine, manufacture or composition of the matter).
Step 2A : Prong One: ( whether a claim recites a judicial exception ?) the claim(s) recite ... a computer method/system/medium implemented asymmetric dual encoder... comprising: a “token embedder layer section” having a first token embedding section associated with a first input and a second token embedding section associated with a second input;
“an encoder layer section” having a first encoder section “configured to receive token embeddings from the first token” embedding section and a second encoder section configured to “receive token embeddings” from the second token embedding section;
“a projection layer configured to receive encodings” from both the first and second encoder sections and to generate a set of projections, wherein the projection layer is “shared by the asymmetric dual encoder system”; and
“an embedding space configured”, “based on the set of projections, to generate a question embedding and an answer embedding”, the question and answer embeddings for use in identifying a set of candidate answers to an input answer...
These limitation(s) recite mental processes and mathematical calculation...since... “a projection layer configured to receive encodings” from both the first and second encoder sections.... [is a high level mathematical calculation(s) (see the current specifications USPGPUB 20240346290 Para(s) 30 and 56-57 for this interpretations...] ...then [APPLY IT] to “generate a set of projections,” wherein the projection layer ... and an embedding space configured , based on the set of projections, to generate a question embedding and an answer embedding for use in identifying a set of candidate answers to an input answer....Thus these limitation(s) recite mental processes and mathematical calculation(s).
--------------Step 2A : Prong Two: (Do the claim(s) recite “additional element(s) that integrate the “Judicial Exception” into “A Practical Application” ? The claim(s) recite additional limitation(s) such as “computer” implemented asymmetric dual encoder.... configured to encode/decode the token embeddings from the token(s) embedding section(s) and ... a projection layer configured to receive encodings from both the first and second encoder sections and to generate a set of projections, wherein the projection layer is shared by the asymmetric dual encoder system; and an embedding space configured, based on the set of projections, to generate a question embedding and an answer embedding”, the question and answer embeddings for use in identifying a set of candidate answers to an input answer...it is noted, the improvement in the abstract idea itself ... but do not integrate the judicial exception into a practical application, i.e., applied the “encode/decode the token embeddings from the token(s) embedding section(s) and ... a projection layer”... to generate a question embedding and an answer embedding for use in identifying a set of candidate answers to an input answer ...
These limitation(s) only recite a generic computer component(s) that only amounts to mere instructions to implement the abstract idea on a computer, and therefore, do not integrate the judicial exception into a practical application. (MPEP 2106.04(d), 2106.05(f)).
Step 2B: (Whether a Claim Amounts to Significantly More) ? The claim(s) recite additional limitation(s) such as ... applied the “computer implemented” and applied the “encode/decode the token embeddings from the token(s) embedding section(s) and ... a projection layer”... to generate a question embedding and an answer embedding for use in identifying a set of candidate answers to an input answer ......These limitation(s) only recite a generic computer component(s) that only amounts to mere instructions to implement the abstract idea on a computer, and therefore, do not amount to significantly more than the abstract idea itself (MPEP 2106.05, 2106.04(d) and 2106.05(f)).
As to the dependent claim(s) 2-12, 14-20 further recite, addition limitation(s) such as, (token(s) embedding sections and encoder(s) section(s) are distinctly parameterized, type(s) of mixed-input source, the first type of input comprises text, and the second type of input does not include text, the second type of input includes at least one of imagery or audio, second type of input includes a structured form, dual encoder system is trained by optimizing contrastive loss with an in-batch sampled soft-max, cosine distance is used as a similarity function for the contrastive loss and projection layer is randomly initialized, etc.,) These limitation(s) only amounts to mere instructions to implement the abstract idea ...and do not include elements that amount to significantly more than the abstract idea and are also rejected under the same rational.
Accordingly, claims 1-21 fail to recite statutory subject matter, as defined in 35 U.S.C. 101.
In addition, Claim(s) 1-12 recite A computer-implemented asymmetric dual encoder “system”. The Specification in USPGPUB 20240346290 A1, paragraph 73 and FIG. 13 illustrates an exemplary method 1300 implementing an asymmetric dual encoder... the asymmetric dual encoder system, the first and second encodings. At block 1312 the method includes generating, by the shared projection layer, a set of projections according to the first and second encodings. And at block 1314 the method includes generating, in an embedding space based on the set of projections, a question embedding and an answer embedding. The question and answer embeddings can then be used in identifying a set of candidate answers to an input answer ...”. Thus, the claim is construed broadly as a computer “Program”, etc. Accordingly, Claim 1-12 fails to recite statutory subject matter, as defined in 35 U.S.C. 101.
Claims Rejection – 35 U.S.C. 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 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.
Claim(s) 1-9, 12-17 and 19-21 rejected under 35 U.S.C. 103 as being unpatentable over Dontcheva et al., (“US 20240134597 A1” filed 10/17/2022 [hereinafter “Dontcheva”], in view of Osuala et al., (“US 20240111794 A1” filed 09/23/2022 [hereinafter “Osuala”].
Independent Claim 1, Dontcheva teaches: A computer-implemented asymmetric dual encoder system, comprising: a token embedding layer section(s) ...., (In Dontcheva Para(s) 10, 78 and Fig. 1, discloses receiving and identifying an initial set of questions asked in the video by parsing a diarized transcript of the video, moreover, Dontcheva teaches: an encoder layer section having a first encoder section configured to receive token embeddings from the first token embedding section and a second encoder section configured to receive token embeddings from the second token embedding section ( In Dontcheva Para 10, i.e., encoding video frames into a frame embedding, and freeform textual query into a query embedding).. a projection layer configured to receive encodings from both the first and second encoder sections and to generate a set of projections, wherein the projection layer is shared by the asymmetric dual encoder system. ( In Dontcheva Para 10, i.e., encoding the text and visual query into a common embedding space); Further, Dontcheva teaches: and an embedding space configured, based on the set of projections, to generate a question embedding and an answer embedding, the question and answer embeddings for use in identifying a set of candidate answers to an input answer (in Dontcheva Para 10, i.e., using a nearest neighbor search to identify and search for video frame embeddings that best match them query embeddings...)
It is noted, Dontcheva discloses computer-implemented asymmetric dual encoder system, comprising: a token embedder layer section having a first token embedding section associated with a first input and a second token embedding section associated with a second input...(as describes herein) However, Dontcheva does not expressly teach .However, the combination of Dontcheva and Osuala teach the limitation said: a token embedder layer section having a first token embedding section and a second token embedding section (in Osuala Para(s) 4, 26-29 and the Abstract, i.e., the embedding generation model may be fine-tuned with additional tasks. respectively...wherein the token embedder layer section having a first token embedding section and a second token embedding section...)
Accordingly, it would have been obvious to one having ordinary skill in the art at the time before the effective filing date of the claimed invention was made to modify Dontcheva’s computer-implemented asymmetric dual encoder system, comprising: a token embedder layer section..., to include a means said a token embedder layer section having a first token embedding section and a second token embedding section, as taught by Osuala; that provides searching a data source using embeddings of vector space; that makes up a large fraction of daily activities in many job roles and companies...(in Osuala Para(s) 1-2). It is noted the KSR ruling recommends references directed to similar subject matter to be combined.
Claim 2, Dontcheva and Osuala further teach: wherein the first and second token embedding sections are distinctly parameterized; (Osuala Para 101, i.e., input is/are question/answer wherein the input ...)
Accordingly, it would have been obvious to one having ordinary skill in the art at the time before the effective filing date of the claimed invention was made to modify Dontcheva’s computer-implemented asymmetric dual encoder system, comprising: a token embedder layer section..., to include a means said ... the first input is a question and the second input is an answer, as taught by Osuala; that provides searching a data source using embeddings of vector space; that makes up a large fraction of daily activities in many job roles and companies...(in Osuala Para(s) 1-2). It is noted the KSR ruling recommends references directed to similar subject matter to be combined.
Claim 3, Dontcheva and Osuala further teach: wherein the first and second token embedding sections are distinctly parameterized; (Osuala Para 101, i.e., muti-task model may receive as input the answer of the query... the layer .... to the two multi-task models 1001 and 1101 whose parameters may be adapted for a next iteration using another pair of question-answer...)
Accordingly, it would have been obvious to one having ordinary skill in the art at the time before the effective filing date of the claimed invention was made to modify Dontcheva’s computer-implemented asymmetric dual encoder system, comprising: a token embedder layer section..., to include a means said ... wherein the first and second token embedding sections are distinctly parameterized, as taught by Osuala; that provides searching a data source using embeddings of vector space; that makes up a large fraction of daily activities in many job roles and companies...(in Osuala Para(s) 1-2). It is noted the KSR ruling recommends references directed to similar subject matter to be combined.
Claim 4, Dontcheva and Osuala further teach: wherein the first and second encoder sections are distinctly parameterized; (Osuala Para 101, i.e., muti-task model may receive as input the answer of the query... the layer .... to the two multi-task models 1001 and 1101 whose parameters may be adapted for a next iteration using another pair of question-answer...)
Accordingly, it would have been obvious to one having ordinary skill in the art at the time before the effective filing date of the claimed invention was made to modify Dontcheva’s computer-implemented asymmetric dual encoder system, comprising: a token embedder layer section..., to include a means said ... wherein the first and second encoder sections are distinctly parameterized, as taught by Osuala; that provides searching a data source using embeddings of vector space; that makes up a large fraction of daily activities in many job roles and companies...(in Osuala Para(s) 1-2). It is noted the KSR ruling recommends references directed to similar subject matter to be combined.
Claim 5, Dontcheva and Osuala further teach: wherein the asymmetric dual encoder system is configured to receive input from a mixed-input source, a first type of input from the mixed-input source to be received by the first token embedding section and a second type of input from the mixed-input source to be received by the second token embedding section. (In Dontcheva Para(s) 10, 78 and Fig. 1, discloses receiving and identifying an initial set of questions asked in the video by parsing a diarized transcript of the video, moreover, Dontcheva teaches: the encoding video frames into a frame embedding, and freeform textual query into a query embedding....i.e., mixed-input source to be received...)
Claim 6, Dontcheva and Osuala further teach: wherein the first type of input comprises text, and the second type of input does not include text. (In Dontcheva Para(s) 10, 78 and Fig. 1, discloses receiving and identifying an initial set of questions asked in the video by parsing a diarized transcript of the video, moreover, Dontcheva teaches: the encoding video frames into a frame embedding, and freeform textual query into a query embedding....i.e., mixed-input source to be received...)
Claim 7, Dontcheva and Osuala further teach: wherein the second type of input includes at least one of imagery or audio. (In Dontcheva Para(s) 10, 78 and Fig. 1, discloses receiving and identifying an initial set of questions asked in the video by parsing a diarized transcript of the video, moreover, Dontcheva teaches: the encoding video frames into a frame embedding, and freeform textual query into a query embedding ...)
Claim 8, Dontcheva and Osuala further teach: wherein the second type of input includes a structured form. (In Dontcheva Para(s) 10, 71, 78 and Fig. 1, discloses receiving and identifying an initial set of questions asked in the video by parsing a diarized transcript of the video, moreover, Dontcheva teaches: the encoding video frames into a frame embedding, and freeform textual query into a query embedding and any application capable of facilitating video editing or playback, such as a stand-alone application, a mobile application, a web application, and/or the like. In some implementations, the application(s) comprises a web application, for example, that is accessible through a web browser, hosted at least partially server-side, and/or the like (structure form).)
Claim 9, Dontcheva and Osuala further teach: wherein the first and second token embedding sections are initialized from a same set of pre-trained parameters, but are fine-tuned separately. (Osuala Para(s) 28 and 101, i.e., muti-task model may receive as input the answer of the query... the layer .... to the two multi-task models 1001 and 1101 whose parameters may be adapted for a next iteration using another pair of question-answer... wherein the pre-trained embedding generation model may be provided. The embedding generation model may, for example, be a deep neural network or another model that predicts embeddings according to the present subject matter and the resulting trained embedding generation model may generalize better and make more accurate predictions than a model for a single task. For example, if the embedding generation model is a deep neural network, with the MTL, the input tokens share the same hidden representations. The present subject matter may provide a unique combination of multi-task learning and token-based embedding generation. .)
Accordingly, it would have been obvious to one having ordinary skill in the art at the time before the effective filing date of the claimed invention was made to modify Dontcheva’s computer-implemented asymmetric dual encoder system, comprising: a token embedder layer section..., to include a means said ... wherein the first and second token embedding sections are initialized from a same set of pre-trained parameters, but are fine-tuned separately, as taught by Osuala; that provides searching a data source using embeddings of vector space; that makes up a large fraction of daily activities in many job roles and companies...(in Osuala Para(s) 1-2). It is noted the KSR ruling recommends references directed to similar subject matter to be combined.
Claim 12, Dontcheva and Osuala further teach: wherein during training the projection layer is randomly initialized. (In Dontcheva Para(s) 10, 78 and Fig. 1, discloses receiving and identifying an initial set of questions asked in the video by parsing a diarized transcript of the video, moreover, Dontcheva teaches: encoding video frames into a frame embedding, and freeform textual query into a query embedding... encoding the text and visual query into a common embedding space... using a nearest neighbor search to identify and search for video frame embeddings that best match them query embeddings...)
Regarding Claim(s) 13-17 (respectively) is/are fully incorporated similar subject of claim(s) 1, 3-4, (3+4) and 9 (respectively) cited above.
Regarding Claim(s) 19-20 (respectively) is/are fully incorporated similar subject of claim(s) 12 and 1 (respectively) cited above.
Regarding Claim 21 is/are fully incorporated similar subject of claim 1 cited above.
Claim(s) 10-11 and 18 rejected under 35 U.S.C. 103 as being unpatentable over Dontcheva et al., (“US 20240134597 A1” filed 10/17/2022 [hereinafter “Dontcheva”], in view of Osuala et al., (“US 20240111794 A1” filed 09/23/2022 [hereinafter “Osuala”] and further in view of Rothberg et al., (“US 20190347523 A1” filed 05/08/2019 [hereinafter “Rothberg”];
Claim 10, Dontcheva and Osuala do not expressly teach, wherein the dual encoder system is trained by optimizing contrastive loss with an in-batch sampled soft-max. However, the combination of Dontcheva and Osuala and Rothberg teach these limitation(s) in Rothberg Para(s) 9-14 and 106, i.e., the dual encoder system is trained by optimizing contrastive loss with an in-batch sampled soft-max. .)
Accordingly, it would have been obvious to one having ordinary skill in the art at the time before the effective filing date of the claimed invention was made to modify Dontcheva and Osuala’s method..., to include a means said ... wherein the dual encoder system is trained by optimizing contrastive loss with an in-batch sampled soft-max, as taught by Rothberg; that provides searching a data source using embeddings of vector space; that makes up a large fraction of daily activities in many job roles and companies...(in Osuala Para(s) 1-2). It is noted the KSR ruling recommends references directed to similar subject matter to be combined.
Claim 11, Dontcheva and Osuala and Rothberg further teach: wherein cosine distance is used as a similarity function for the contrastive loss. (in Rothberg Para(s) 9-14 and 106, i.e., the dual encoder system is trained by optimizing contrastive loss with cosine distance is used as a similarity function for the contrastive loss. .)
Accordingly, it would have been obvious to one having ordinary skill in the art at the time before the effective filing date of the claimed invention was made to modify Dontcheva and Osuala’s method..., to include a means said ... wherein cosine distance is used as a similarity function for the contrastive loss, as taught by Rothberg; that provides searching a data source using embeddings of vector space; that makes up a large fraction of daily activities in many job roles and companies...(in Osuala Para(s) 1-2). It is noted the KSR ruling recommends references directed to similar subject matter to be combined.
Regarding Claim 18 is/are fully incorporated similar subject of claim 10 cited above.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
Grail (“ US 20210256069 A1” filed 09/09/2020, relates to a question answering system includes: a first encoder module configured to receive a question, the question including a first plurality of words, and encode the question into a first vector representation; a second encoder module configured to encode a document into a second vector representation, the document including a second plurality of words; a first reading module configured to generate a third vector representation based on the first and second vector representations; a first reformulation module configured to generate a first reformulated vector representation based on the first vector representation; a second reading module configured to generate a fifth vector representation based on the second vector representation and the first reformulated vector representation; a second reformulation module configured to generate a second reformulated vector representation based on first reformulated vector representation; and an answer module configured to determine an answer to the question based on the second reformulated vector representation... [the Abstract].
Liu et al., NPL (“Point2Token: A Multi-Tagging Answer Retrieval Framework for Question Answering” Published 2021 by IEEE, 5 pages, describing, question answering plays a crucial role in the chatbot systems, in which it retrieves the answer from the given context and return the predicted span as a result to users. Previous work mostly modelled this task as a multi-classification problem. However, the models cannot gain a promising result due to the scarcity of the probability distribution over the whole given context. In this paper, we propose a novel approach to solve the problem mentioned above. We model the question answering task as a multiple binary classification problem and introduce PointerNet in our model decoder to predict whether it belongs to a start or end position in each token within context. The experimental results on a well-studied dataset show that our model outperlorms the baseline models, which proves our model effectiveness. [The Abstract].
Olabanji Shonibare NPL (“ASBERT: Siamese and Triplet network embedding for open question answering” Published 2021, 10 pages, describing answer selection (AS) is an essential subtask in the field of natural language processing with an objective to identify the most likely answer to a given question from a corpus containing candidate answer sentences. A common approach to address the AS problem is to generate an embedding for each candidate sentence and query. Then, select the sentence whose vector representation is closest to the query’s. A key drawback is the low quality of the embeddings, hitherto, based on its performance on AS benchmark datasets. In this work, we present ASBERT, a framework built on the BERT architecture that employs Siamese and Triplet neural networks to learn an encoding function that maps a text to a fixed-size vector in an embedded space. The notion of distance between two points in this space connotes similarity in meaning between two texts. Experimental results on the WikiQA and TrecQA datasets demonstrate that our proposed approach outperforms many state-of-the-art baseline methods.. [The Abstract].
Any inquiry concerning this communication or earlier communications from the examiner should be directed to QUOC A TRAN whose telephone number is (571)272-8664. The examiner can normally be reached Monday-Friday 9am-5pm 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, Cesar Paula can be reached at 571-272-4128. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of published or unpublished applications may be obtained from Patent Center. Unpublished application information in Patent Center is available to registered users. To file and manage patent submissions in Patent Center, visit: https://patentcenter.uspto.gov. Visit https://www.uspto.gov/patents/apply/patent-center for more information about Patent Center and https://www.uspto.gov/patents/docx for information about filing in DOCX format. For additional questions, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/QUOC A TRAN/Primary Examiner, Art Unit 2145
Prosecution Insights