DETAILED ACTION
This action is in response to the Applicant Response filed 21 May 2026 for application 18/135,958 filed 18 April 2023.
Claim(s) 24-26, 2831, 35, 38, 41-42 is/are currently amended.
Claim(s) 44-46 is/are new.
Claim(s) 1-23, 27, 34, 36 is/are cancelled.
Claim(s) 24-26, 28-33, 35, 37-46 is/are pending.
Claim(s) 24-26, 28-33, 35, 37-46 is/are rejected.
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 Arguments
Applicant's arguments regarding the objections to the claims have been fully considered and, in light of the amendments to the claims, are persuasive. However, in light of the amendments to the claims, new claim objections have arisen, as noted below.
Applicant's arguments regarding the 35 U.S.C. 112(a) rejection(s) of claim(s) 24-30 have been fully considered and, in light of the amendments to the claims, are persuasive. The 35 U.S.C. 112(a) rejection(s) of claim(s) 24-30 has/have been withdrawn.
Applicant's arguments regarding the 35 U.S.C. 112(b) rejection(s) of claim(s) 24-30 have been fully considered and, in light of the amendments to the claims, are persuasive. The 35 U.S.C. 112(b) rejection(s) of claim(s) 24-30 has/have been withdrawn.
Applicant’s arguments regarding the 35 U.S.C. 101 rejection of claims 24-26, 28-33, 35, 37-46 have been fully considered but are not persuasive.
Applicant first argues that the claims recite various circuits and therefore do not recite a mental process. Examiner respectfully disagrees. The circuits recited in the claims are recited at a high-level of generality such that they are simply generic computer components performing generic computer functions of executing instructions on the computers and amount to no more than mere instructions to apply the exception using generic computer components (MPEP 2106.05(b)).
Applicant argues that in light of Desjardins, the claim is eligible. Examiner respectfully disagrees. Desjardins includes an improvement related to catastrophic forgetting while the claims in the instant application only recite machine learning at a high level to perform an abstract idea, as detailed below. While Desjardins provides a specific training strategy that allows models to preserve performance on earlier tasks even as it learns new ones, the current claims merely employ generic machine learning at a high level and recite known mathematical steps of generic machine learning function using generic computer components.
Therefore, claims 24-26, 28-33, 35, 37-46 stand rejected under 35 U.S.C. 101.
Applicant’s arguments regarding the 35 U.S.C. 102 and/or 35 U.S.C. 103 rejections of the claims are based on the newly amended subject matter. All arguments are addressed in the 35 U.S.C. 102 and/or 35 U.S.C. 103 rejections of the claims below.
Claim Objections
Claim(s) 29, 44 is/are objected to because of the following informalities:
Claim 29, line 2, wherein compression circuit should read “wherein the compression circuit”
Claim 29, line 3, compress the values should read “compress the value” [“s” removed from values]
Claim 44, line 2, the comma and word “and” at the end of the line should be replaced with a period
Appropriate correction is required.
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) 24-26, 28-33, 35, 37-46 is/are rejected under 35 U.S.C. 101, because the claim(s) is/are directed to an abstract idea, and because the claim elements, whether considered individually or in combination, do not amount to significantly more than the abstract idea, see Alice Corporation Pty. Ltd. V. CLS Bank International et al., 573 US 208 (2014).
Regarding claim 24, the claim is rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
Step 1 Analysis: Claim 24 is directed to a(n) apparatus, which is directed to a machine, one of the statutory categories.
Step 2A Prong One Analysis: The claim recites a(n) apparatus for deep learning.
The limitation of decompose an element in an input feature map of a deep learning operation into two input elements, the element in the input feature map having a first precision, the two input elements having a second precision that is lower than the first precision, as drafted, is a process that, under its broadest reasonable interpretation, covers a mental process. The limitation is directed to observation, evaluation, judgment and opinion and is a process capable of being performed by a human mentally or using pen and paper.
The limitation of decompose a weight of the deep learning operation into two weight elements, the weight associated with the element in the input feature map, the weight having the first precision, the two weight elements having the second precision, as drafted, is a process that, under its broadest reasonable interpretation, covers a mental process. The limitation is directed to observation, evaluation, judgment and opinion and is a process capable of being performed by a human mentally or using pen and paper.
The limitation of perform a computation using the two input elements and the two weight elements, as drafted, is a process that, under its broadest reasonable interpretation, covers a mathematical concept. The limitation encompasses calculate a value.
The limitation of skip one or more computations of one or more zero-valued elements in the input feature map by the multiply-accumulation circuit, as drafted, is a process that, under its broadest reasonable interpretation, covers a mental process. The limitation is directed to observation, evaluation, judgment and opinion and is a process capable of being performed by a human mentally or using pen and paper.
The limitation of compress values computed by the multiply- accumulate circuit, wherein the compressed values are output activations of the deep learning operation and are to be processed in a next deep learning operation, as drafted, is a process that, under its broadest reasonable interpretation, covers a mental process. The limitation is directed to observation, evaluation, judgment and opinion and is a process capable of being performed by a human mentally or using pen and paper.
If a claim limitation, under its broadest reasonable interpretation, covers performance of the limitation in the mind, then it falls within the "Mental Processes" grouping. If a claim limitation, under its broadest reasonable interpretation, covers performance of mathematical concepts, then it falls within the "Mathematical Concepts" grouping. Accordingly, the claim recites an abstract idea.
Step 2A Prong Two Analysis: With respect to the abstract idea, the judicial exception is not integrated into a practical application.
The claim recites additional element(s) – apparatus, mixed precision circuit, multiply-accumulation circuit, sparsity circuit, compression circuit. The additional element(s) is/are recited at a high-level of generality (i.e., as generic computer components performing generic computer functions of executing instructions on the computers) such that it amounts to no more than mere instructions to apply the exception using generic computer components (MPEP 2106.05(b)).
The claim recites additional element(s) – deep learning. The additional element(s) is/are recited at a high-level of generality such that it amounts to no more than indicating a field of use or technological environment in which to apply the judicial exception (MPEP 2106.05(h)).
Accordingly, the additional element(s) do(es) not integrate the abstract idea into a practical application because the additional element(s) do(es) not impose any meaningful limits on practicing the abstract idea, and, therefore, the claim is directed to an abstract idea.
Step 2B Analysis: The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception. As discussed above with respect to the integration of the abstract idea into a practical application, the additional element(s) of:
apparatus, mixed precision circuit, multiply-accumulation circuit, sparsity circuit, compression circuit amount(s) to no more than mere instructions to apply the exception using generic computer components (MPEP 2106.05(b))
deep learning amount(s) to no more than indicating a field of use or technological environment in which to apply the judicial exception (MPEP 2106.05(h))
The additional element(s) do(es) not provide an inventive concept, and, therefore, the claim is not patent eligible.
Regarding claim 25, the claim is rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
Step 1 Analysis: Claim 25 is directed to a(n) apparatus, which is directed to a machine, one of the statutory categories.
Step 2A Prong One Analysis: The claim recites a(n) apparatus for deep learning.
The limitation of compress a plurality of elements in the input feature map, the plurality of elements comprising the element, as drafted, is a process that, under its broadest reasonable interpretation, covers a mental process. The limitation is directed to observation, evaluation, judgment and opinion and is a process capable of being performed by a human mentally or using pen and paper.
If a claim limitation, under its broadest reasonable interpretation, covers performance of the limitation in the mind, then it falls within the "Mental Processes" grouping. Accordingly, the claim recites an abstract idea.
Step 2A Prong Two Analysis: With respect to the abstract idea, the judicial exception is not integrated
into a practical application. The claim does not recite any additional elements which integrate the
abstract idea into a practical application and, therefore, does not impose any meaningful limits on
practicing the abstract idea. Therefore, the claim is directed to an abstract idea.
Step 2B Analysis: The claim does not include additional elements that are sufficient to amount to
significantly more than the judicial exception. As discussed above with respect to the integration of the
abstract idea into a practical application, the claim does not recite any additional elements which
provide an inventive concept, and, therefore, the claim is not patent eligible.
Regarding claim 26, the claim is rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
Step 1 Analysis: Claim 26 is directed to a(n) apparatus, which is directed to a machine, one of the statutory categories.
Step 2A Prong One Analysis: The claim recites a(n) apparatus for deep learning.
The limitation of compress a plurality of weights of the deep learning operation, the plurality of weights comprising the weight, as drafted, is a process that, under its broadest reasonable interpretation, covers a mental process. The limitation is directed to observation, evaluation, judgment and opinion and is a process capable of being performed by a human mentally or using pen and paper.
If a claim limitation, under its broadest reasonable interpretation, covers performance of the limitation in the mind, then it falls within the "Mental Processes" grouping. Accordingly, the claim recites an abstract idea.
Step 2A Prong Two Analysis: With respect to the abstract idea, the judicial exception is not integrated
into a practical application. The claim does not recite any additional elements which integrate the
abstract idea into a practical application and, therefore, does not impose any meaningful limits on
practicing the abstract idea. Therefore, the claim is directed to an abstract idea.
Step 2B Analysis: The claim does not include additional elements that are sufficient to amount to
significantly more than the judicial exception. As discussed above with respect to the integration of the
abstract idea into a practical application, the claim does not recite any additional elements which
provide an inventive concept, and, therefore, the claim is not patent eligible.
Regarding claim 28, the claim is rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
Step 1 Analysis: Claim 28 is directed to a(n) apparatus, which is directed to a machine, one of the statutory categories.
Step 2A Prong One Analysis: The claim recites a(n) apparatus for deep learning.
The limitation of decompose a value computed by the multiply-accumulate circuit into two elements having the second precision, as drafted, is a process that, under its broadest reasonable interpretation, covers a mental process. The limitation is directed to observation, evaluation, judgment and opinion and is a process capable of being performed by a human mentally or using pen and paper.
If a claim limitation, under its broadest reasonable interpretation, covers performance of the limitation in the mind, then it falls within the "Mental Processes" grouping. Accordingly, the claim recites an abstract idea.
Step 2A Prong Two Analysis: With respect to the abstract idea, the judicial exception is not integrated
into a practical application. The claim does not recite any additional elements which integrate the
abstract idea into a practical application and, therefore, does not impose any meaningful limits on
practicing the abstract idea. Therefore, the claim is directed to an abstract idea.
Step 2B Analysis: The claim does not include additional elements that are sufficient to amount to
significantly more than the judicial exception. As discussed above with respect to the integration of the
abstract idea into a practical application, the claim does not recite any additional elements which
provide an inventive concept, and, therefore, the claim is not patent eligible.
Regarding claim 29, the claim is rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
Step 1 Analysis: Claim 29 is directed to a(n) apparatus, which is directed to a machine, one of the statutory categories.
Step 2A Prong One Analysis: The claim recites a(n) apparatus for deep learning.
The limitation of compress the values by setting one of the two elements to zero, as drafted, is a process that, under its broadest reasonable interpretation, covers a mental process. The limitation is directed to observation, evaluation, judgment and opinion and is a process capable of being performed by a human mentally or using pen and paper.
If a claim limitation, under its broadest reasonable interpretation, covers performance of the limitation in the mind, then it falls within the "Mental Processes" grouping. Accordingly, the claim recites an abstract idea.
Step 2A Prong Two Analysis: With respect to the abstract idea, the judicial exception is not integrated
into a practical application. The claim does not recite any additional elements which integrate the
abstract idea into a practical application and, therefore, does not impose any meaningful limits on
practicing the abstract idea. Therefore, the claim is directed to an abstract idea.
Step 2B Analysis: The claim does not include additional elements that are sufficient to amount to
significantly more than the judicial exception. As discussed above with respect to the integration of the
abstract idea into a practical application, the claim does not recite any additional elements which
provide an inventive concept, and, therefore, the claim is not patent eligible.
Regarding claim 30, the claim is rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
Step 1 Analysis: Claim 30 is directed to a(n) apparatus, which is directed to a machine, one of the statutory categories.
Step 2A Prong One Analysis: The claim recites a(n) apparatus for deep learning.
The limitation of compute the element in the input feature map based on an activation function, as drafted, is a process that, under its broadest reasonable interpretation, covers a mathematical concept. The limitation encompasses calculate a value.
If a claim limitation, under its broadest reasonable interpretation, covers performance of mathematical concepts, then it falls within the "Mathematical Concepts" grouping. Accordingly, the claim recites an abstract idea.
Step 2A Prong Two Analysis: With respect to the abstract idea, the judicial exception is not integrated into a practical application.
The claim recites additional element(s) – additional circuit. The additional element(s) is/are recited at a high-level of generality (i.e., as generic computer components performing generic computer functions of executing instructions on the computers) such that it amounts to no more than mere instructions to apply the exception using generic computer components (MPEP 2106.05(b)).
Accordingly, the additional element(s) do(es) not integrate the abstract idea into a practical application because the additional element(s) do(es) not impose any meaningful limits on practicing the abstract idea, and, therefore, the claim is directed to an abstract idea.
Step 2B Analysis: The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception. As discussed above with respect to the integration of the abstract idea into a practical application, the additional element(s) of:
additional circuit amount(s) to no more than mere instructions to apply the exception using generic computer components (MPEP 2106.05(b))
The additional element(s) do(es) not provide an inventive concept, and, therefore, the claim is not patent eligible.
Regarding claim 31, the claim is rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
Step 1 Analysis: Claim 31 is directed to a method, which is directed to a process, one of the statutory categories.
Step 2A Prong One Analysis: The claim recites a(n) method for deep learning.
The limitation of decomposing an element in an input feature map of a deep learning operation into two input elements, the element in the input feature map having a first precision, the two input elements having a second precision that is lower than the first precision, the input feature map further comprising one or more zero-valued elements, as drafted, is a process that, under its broadest reasonable interpretation, covers a mental process. The limitation is directed to observation, evaluation, judgment and opinion and is a process capable of being performed by a human mentally or using pen and paper.
The limitation of decomposing a weight of the deep learning operation into two weight elements, the weight associated with the element in the input feature map, the weight having the first precision, the two weight elements having the second precision, as drafted, is a process that, under its broadest reasonable interpretation, covers a mental process. The limitation is directed to observation, evaluation, judgment and opinion and is a process capable of being performed by a human mentally or using pen and paper.
The limitation of performing a computation using the two input elements and the two weight elements to compute a value, as drafted, is a process that, under its broadest reasonable interpretation, covers a mathematical concept. The limitation encompasses calculating a value.
The limitation of skipping one or more computations of the one or more zero-valued elements, as drafted, is a process that, under its broadest reasonable interpretation, covers a mental process. The limitation is directed to observation, evaluation, judgment and opinion and is a process capable of being performed by a human mentally or using pen and paper.
The limitation of compressing the value, wherein the compressed value is an output activation of the deep learning operation and is to be processed in a next deep learning operation, as drafted, is a process that, under its broadest reasonable interpretation, covers a mental process. The limitation is directed to observation, evaluation, judgment and opinion and is a process capable of being performed by a human mentally or using pen and paper.
If a claim limitation, under its broadest reasonable interpretation, covers performance of the limitation in the mind, then it falls within the "Mental Processes" grouping. If a claim limitation, under its broadest reasonable interpretation, covers performance of mathematical concepts, then it falls within the "Mathematical Concepts" grouping. Accordingly, the claim recites an abstract idea.
Step 2A Prong Two Analysis: With respect to the abstract idea, the judicial exception is not integrated into a practical application.
The claim recites additional element(s) – deep learning. The additional element(s) is/are recited at a high-level of generality such that it amounts to no more than indicating a field of use or technological environment in which to apply the judicial exception (MPEP 2106.05(h)).
Accordingly, the additional element(s) do(es) not integrate the abstract idea into a practical application because the additional element(s) do(es) not impose any meaningful limits on practicing the abstract idea, and, therefore, the claim is directed to an abstract idea.
Step 2B Analysis: The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception. As discussed above with respect to the integration of the abstract idea into a practical application, the additional element(s) of:
deep learning amount(s) to no more than indicating a field of use or technological environment in which to apply the judicial exception (MPEP 2106.05(h))
The additional element(s) do(es) not provide an inventive concept, and, therefore, the claim is not patent eligible.
Regarding claim 32, the claim is rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
Step 1 Analysis: Claim 32 is directed to a method, which is directed to a process, one of the statutory categories.
Step 2A Prong One Analysis: The claim recites a(n) method for deep learning.
The limitation of compressing the input feature map, as drafted, is a process that, under its broadest reasonable interpretation, covers a mental process. The limitation is directed to observation, evaluation, judgment and opinion and is a process capable of being performed by a human mentally or using pen and paper.
If a claim limitation, under its broadest reasonable interpretation, covers performance of the limitation in the mind, then it falls within the "Mental Processes" grouping. Accordingly, the claim recites an abstract idea.
Step 2A Prong Two Analysis: With respect to the abstract idea, the judicial exception is not integrated into a practical application.
The claim recites additional element(s) – memory. The additional element(s) is/are recited at a high-level of generality (i.e., as generic computer components performing generic computer functions of executing instructions on the computers) such that it amounts to no more than mere instructions to apply the exception using generic computer components (MPEP 2106.05(b)).
The claim recites storing a compressed version of the input feature map in a memory, which is simply storing data recited at a high level of generality. This is nothing more than insignificant extra-solution activity (MPEP 2106.05(g)).
Accordingly, the additional element(s) do(es) not integrate the abstract idea into a practical application because the additional element(s) do(es) not impose any meaningful limits on practicing the abstract idea, and, therefore, the claim is directed to an abstract idea.
Step 2B Analysis: The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception. As discussed above with respect to the integration of the abstract idea into a practical application, the additional element(s) of:
memory amount(s) to no more than mere instructions to apply the exception using generic computer components (MPEP 2106.05(b))
storing data amount(s) to no more than insignificant extra-solution activity (MPEP 2106.05(g)), wherein the insignificant extra-solution activity is the well-understood routine and conventional activit(y/ies) of storing and retrieving information in memory (MPEP 2016.05(d))
The additional element(s) do(es) not provide an inventive concept, and, therefore, the claim is not patent eligible.
Regarding claim 33, the claim is rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
Step 1 Analysis: Claim 33 is directed to a method, which is directed to a process, one of the statutory categories.
Step 2A Prong One Analysis: The claim recites a(n) method for deep learning.
The limitation of compressing a plurality of weights of the deep learning operation, the plurality of weights comprising the weight, as drafted, is a process that, under its broadest reasonable interpretation, covers a mental process. The limitation is directed to observation, evaluation, judgment and opinion and is a process capable of being performed by a human mentally or using pen and paper.
If a claim limitation, under its broadest reasonable interpretation, covers performance of the limitation in the mind, then it falls within the "Mental Processes" grouping. Accordingly, the claim recites an abstract idea.
Step 2A Prong Two Analysis: With respect to the abstract idea, the judicial exception is not integrated
into a practical application. The claim does not recite any additional elements which integrate the
abstract idea into a practical application and, therefore, does not impose any meaningful limits on
practicing the abstract idea. Therefore, the claim is directed to an abstract idea.
Step 2B Analysis: The claim does not include additional elements that are sufficient to amount to
significantly more than the judicial exception. As discussed above with respect to the integration of the
abstract idea into a practical application, the claim does not recite any additional elements which
provide an inventive concept, and, therefore, the claim is not patent eligible.
Regarding claim 35, the claim is rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
Step 1 Analysis: Claim 35 is directed to a method, which is directed to a process, one of the statutory categories.
Step 2A Prong One Analysis: The claim recites a(n) method for deep learning.
The limitation of decomposing the value into two elements having the second precision, as drafted, is a process that, under its broadest reasonable interpretation, covers a mental process. The limitation is directed to observation, evaluation, judgment and opinion and is a process capable of being performed by a human mentally or using pen and paper.
If a claim limitation, under its broadest reasonable interpretation, covers performance of the limitation in the mind, then it falls within the "Mental Processes" grouping. Accordingly, the claim recites an abstract idea.
Step 2A Prong Two Analysis: With respect to the abstract idea, the judicial exception is not integrated
into a practical application. The claim does not recite any additional elements which integrate the
abstract idea into a practical application and, therefore, does not impose any meaningful limits on
practicing the abstract idea. Therefore, the claim is directed to an abstract idea.
Step 2B Analysis: The claim does not include additional elements that are sufficient to amount to
significantly more than the judicial exception. As discussed above with respect to the integration of the
abstract idea into a practical application, the claim does not recite any additional elements which
provide an inventive concept, and, therefore, the claim is not patent eligible.
Regarding claim 37, the claim is rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
Step 1 Analysis: Claim 37 is directed to a method, which is directed to a process, one of the statutory categories.
Step 2A Prong One Analysis: The claim recites a(n) method for deep learning.
The limitation of computing the element in the input feature map based on an activation function, as drafted, is a process that, under its broadest reasonable interpretation, covers a mathematical concept. The limitation encompasses calculating a value.
If a claim limitation, under its broadest reasonable interpretation, covers performance of mathematical concepts, then it falls within the "Mathematical Concepts" grouping. Accordingly, the claim recites an abstract idea.
Step 2A Prong Two Analysis: With respect to the abstract idea, the judicial exception is not integrated
into a practical application. The claim does not recite any additional elements which integrate the
abstract idea into a practical application and, therefore, does not impose any meaningful limits on
practicing the abstract idea. Therefore, the claim is directed to an abstract idea.
Step 2B Analysis: The claim does not include additional elements that are sufficient to amount to
significantly more than the judicial exception. As discussed above with respect to the integration of the
abstract idea into a practical application, the claim does not recite any additional elements which
provide an inventive concept, and, therefore, the claim is not patent eligible.
Regarding claim 38, the claim is rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
Step 1 Analysis: Claim 38 is directed to computer-readable media, which is directed to an article of manufacture, one of the statutory categories.
Step 2A Prong One Analysis: The claim recites a(n) one or more non-transitory computer-readable media.
The limitation of decomposing an element in an input feature map of a deep learning operation into two input elements, the element in the input feature map having a first precision, the two input elements having a second precision that is lower than the first precision, the input feature map further comprising one or more zero-valued elements, as drafted, is a process that, under its broadest reasonable interpretation, covers a mental process. The limitation is directed to observation, evaluation, judgment and opinion and is a process capable of being performed by a human mentally or using pen and paper.
The limitation of decomposing a weight of the deep learning operation into two weight elements, the weight associated with the element in the input feature map, the weight having the first precision, the two weight elements having the second precision, as drafted, is a process that, under its broadest reasonable interpretation, covers a mental process. The limitation is directed to observation, evaluation, judgment and opinion and is a process capable of being performed by a human mentally or using pen and paper.
The limitation of performing a computation using the two input elements and the two weight elements to compute a value, as drafted, is a process that, under its broadest reasonable interpretation, covers a mathematical concept. The limitation encompasses calculating a value.
The limitation of skipping one or more computations of the one or more zero-valued elements, as drafted, is a process that, under its broadest reasonable interpretation, covers a mental process. The limitation is directed to observation, evaluation, judgment and opinion and is a process capable of being performed by a human mentally or using pen and paper.
The limitation of compressing the value, wherein the compressed value is an output activation of the deep learning operation and is to be processed in a next deep learning operation, as drafted, is a process that, under its broadest reasonable interpretation, covers a mental process. The limitation is directed to observation, evaluation, judgment and opinion and is a process capable of being performed by a human mentally or using pen and paper.
If a claim limitation, under its broadest reasonable interpretation, covers performance of the limitation in the mind, then it falls within the "Mental Processes" grouping. If a claim limitation, under its broadest reasonable interpretation, covers performance of mathematical concepts, then it falls within the "Mathematical Concepts" grouping. Accordingly, the claim recites an abstract idea.
Step 2A Prong Two Analysis: With respect to the abstract idea, the judicial exception is not integrated into a practical application.
The claim recites additional element(s) – one or more ... computer-readable media, instructions. The additional element(s) is/are recited at a high-level of generality (i.e., as generic computer components performing generic computer functions of executing instructions on the computers) such that it amounts to no more than mere instructions to apply the exception using generic computer components (MPEP 2106.05(b)).
The claim recites additional element(s) – deep learning. The additional element(s) is/are recited at a high-level of generality such that it amounts to no more than indicating a field of use or technological environment in which to apply the judicial exception (MPEP 2106.05(h)).
Accordingly, the additional element(s) do(es) not integrate the abstract idea into a practical application because the additional element(s) do(es) not impose any meaningful limits on practicing the abstract idea, and, therefore, the claim is directed to an abstract idea.
Step 2B Analysis: The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception. As discussed above with respect to the integration of the abstract idea into a practical application, the additional element(s) of:
one or more ... computer-readable media, instructions amount(s) to no more than mere instructions to apply the exception using generic computer components (MPEP 2106.05(b))
deep learning amount(s) to no more than indicating a field of use or technological environment in which to apply the judicial exception (MPEP 2106.05(h))
The additional element(s) do(es) not provide an inventive concept, and, therefore, the claim is not patent eligible.
Regarding claim 39, the claim is rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
Step 1 Analysis: Claim 39 is directed to computer-readable media, which is directed to an article of manufacture, one of the statutory categories.
Step 2A Prong One Analysis: The claim recites a(n) one or more non-transitory computer-readable media.
The limitation of compressing the input feature map, as drafted, is a process that, under its broadest reasonable interpretation, covers a mental process. The limitation is directed to observation, evaluation, judgment and opinion and is a process capable of being performed by a human mentally or using pen and paper.
If a claim limitation, under its broadest reasonable interpretation, covers performance of the limitation in the mind, then it falls within the "Mental Processes" grouping. Accordingly, the claim recites an abstract idea.
Step 2A Prong Two Analysis: With respect to the abstract idea, the judicial exception is not integrated into a practical application.
The claim recites additional element(s) – memory. The additional element(s) is/are recited at a high-level of generality (i.e., as generic computer components performing generic computer functions of executing instructions on the computers) such that it amounts to no more than mere instructions to apply the exception using generic computer components (MPEP 2106.05(b)).
The claim recites storing a compressed version of the input feature map in a memory, which is simply storing data recited at a high level of generality. This is nothing more than insignificant extra-solution activity (MPEP 2106.05(g)).
Accordingly, the additional element(s) do(es) not integrate the abstract idea into a practical application because the additional element(s) do(es) not impose any meaningful limits on practicing the abstract idea, and, therefore, the claim is directed to an abstract idea.
Step 2B Analysis: The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception. As discussed above with respect to the integration of the abstract idea into a practical application, the additional element(s) of:
memory amount(s) to no more than mere instructions to apply the exception using generic computer components (MPEP 2106.05(b))
storing data amount(s) to no more than insignificant extra-solution activity (MPEP 2106.05(g)), wherein the insignificant extra-solution activity is the well-understood routine and conventional activit(y/ies) of storing and retrieving information in memory (MPEP 2016.05(d))
The additional element(s) do(es) not provide an inventive concept, and, therefore, the claim is not patent eligible.
Regarding claim 40, the claim is rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
Step 1 Analysis: Claim 40 is directed to computer-readable media, which is directed to an article of manufacture, one of the statutory categories.
Step 2A Prong One Analysis: The claim recites a(n) one or more non-transitory computer-readable media.
The limitation of compressing a plurality of weights of the deep learning operation, the plurality of weights comprising the weight, as drafted, is a process that, under its broadest reasonable interpretation, covers a mental process. The limitation is directed to observation, evaluation, judgment and opinion and is a process capable of being performed by a human mentally or using pen and paper.
If a claim limitation, under its broadest reasonable interpretation, covers performance of the limitation in the mind, then it falls within the "Mental Processes" grouping. Accordingly, the claim recites an abstract idea.
Step 2A Prong Two Analysis: With respect to the abstract idea, the judicial exception is not integrated
into a practical application. The claim does not recite any additional elements which integrate the
abstract idea into a practical application and, therefore, does not impose any meaningful limits on
practicing the abstract idea. Therefore, the claim is directed to an abstract idea.
Step 2B Analysis: The claim does not include additional elements that are sufficient to amount to
significantly more than the judicial exception. As discussed above with respect to the integration of the
abstract idea into a practical application, the claim does not recite any additional elements which
provide an inventive concept, and, therefore, the claim is not patent eligible.
Regarding claim 41, the claim is rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
Step 1 Analysis: Claim 41 is directed to computer-readable media, which is directed to an article of manufacture, one of the statutory categories.
Step 2A Prong One Analysis: The claim recites a(n) one or more non-transitory computer-readable media.
The limitation of generating an output activation of the deep learning operation from the computation, as drafted, is a process that, under its broadest reasonable interpretation, covers a mathematical concept. The limitation encompasses calculating a value.
If a claim limitation, under its broadest reasonable interpretation, covers performance of mathematical concepts, then it falls within the "Mathematical Concepts" grouping. Accordingly, the claim recites an abstract idea.
Step 2A Prong Two Analysis: With respect to the abstract idea, the judicial exception is not integrated
into a practical application. The claim does not recite any additional elements which integrate the
abstract idea into a practical application and, therefore, does not impose any meaningful limits on
practicing the abstract idea. Therefore, the claim is directed to an abstract idea.
Step 2B Analysis: The claim does not include additional elements that are sufficient to amount to
significantly more than the judicial exception. As discussed above with respect to the integration of the
abstract idea into a practical application, the claim does not recite any additional elements which
provide an inventive concept, and, therefore, the claim is not patent eligible.
Regarding claim 42, the claim is rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
Step 1 Analysis: Claim 42 is directed to computer-readable media, which is directed to an article of manufacture, one of the statutory categories.
Step 2A Prong One Analysis: The claim recites a(n) one or more non-transitory computer-readable media.
The limitation of decomposing the output activation into two output elements having the second precision, as drafted, is a process that, under its broadest reasonable interpretation, covers a mental process. The limitation is directed to observation, evaluation, judgment and opinion and is a process capable of being performed by a human mentally or using pen and paper.
If a claim limitation, under its broadest reasonable interpretation, covers performance of the limitation in the mind, then it falls within the "Mental Processes" grouping. Accordingly, the claim recites an abstract idea.
Step 2A Prong Two Analysis: With respect to the abstract idea, the judicial exception is not integrated
into a practical application. The claim does not recite any additional elements which integrate the
abstract idea into a practical application and, therefore, does not impose any meaningful limits on
practicing the abstract idea. Therefore, the claim is directed to an abstract idea.
Step 2B Analysis: The claim does not include additional elements that are sufficient to amount to
significantly more than the judicial exception. As discussed above with respect to the integration of the
abstract idea into a practical application, the claim does not recite any additional elements which
provide an inventive concept, and, therefore, the claim is not patent eligible.
Regarding claim 43, the claim is rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
Step 1 Analysis: Claim 43 is directed to computer-readable media, which is directed to an article of manufacture, one of the statutory categories.
Step 2A Prong One Analysis: The claim recites a(n) one or more non-transitory computer-readable media.
The limitation of computing the element in the input feature map based on an activation function, as drafted, is a process that, under its broadest reasonable interpretation, covers a mathematical concept. The limitation encompasses calculating a value.
If a claim limitation, under its broadest reasonable interpretation, covers performance of mathematical concepts, then it falls within the "Mathematical Concepts" grouping. Accordingly, the claim recites an abstract idea.
Step 2A Prong Two Analysis: With respect to the abstract idea, the judicial exception is not integrated
into a practical application. The claim does not recite any additional elements which integrate the
abstract idea into a practical application and, therefore, does not impose any meaningful limits on
practicing the abstract idea. Therefore, the claim is directed to an abstract idea.
Step 2B Analysis: The claim does not include additional elements that are sufficient to amount to
significantly more than the judicial exception. As discussed above with respect to the integration of the
abstract idea into a practical application, the claim does not recite any additional elements which
provide an inventive concept, and, therefore, the claim is not patent eligible.
Regarding claim 44, the claim is rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
Step 1 Analysis: Claim 44 is directed to a(n) apparatus, which is directed to a machine, one of the statutory categories.
Step 2A Prong One Analysis: The claim recites a(n) apparatus for deep learning. The Step 2A Prong One Analysis for claim 29 is applicable here since claim 44 carries out the apparatus of claim 29 but for the recitation of additional element(s) of wherein the one of the two elements corresponds to most significant bits of the value.
Step 2A Prong Two Analysis: With respect to the abstract idea, the judicial exception is not integrated into a practical application. In particular, the claim recites additional information regarding the data and the element(s) do(es) not apply the exception in a meaningful way (MPEP 2106.05(e)). Accordingly, the additional element(s) do(es) not integrate the abstract idea into a practical application because the additional element(s) do(es) not impose any meaningful limits on practicing the abstract idea, and, therefore, the claim is directed to an abstract idea.
Step 2B Analysis: The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception. As discussed above with respect to the integration of the abstract idea into a practical application, the additional element(s) of additional information regarding the data do(es) not apply the exception in a meaningful way (MPEP 2106.05(e)). Not applying the exception in a meaningful way does not provide an inventive concept, and, therefore, the claim is not patent eligible.
Regarding claim 45, the claim is rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
Step 1 Analysis: Claim 45 is directed to a method, which is directed to a process, one of the statutory categories.
Step 2A Prong One Analysis: The claim recites a(n) method for deep learning.
The limitation of wherein compressing the value comprises setting one of the two elements to zero, as drafted, is a process that, under its broadest reasonable interpretation, covers a mental process. The limitation is directed to observation, evaluation, judgment and opinion and is a process capable of being performed by a human mentally or using pen and paper.
If a claim limitation, under its broadest reasonable interpretation, covers performance of the limitation in the mind, then it falls within the "Mental Processes" grouping. Accordingly, the claim recites an abstract idea.
Step 2A Prong Two Analysis: With respect to the abstract idea, the judicial exception is not integrated into a practical application.
The claim recites wherein the one of the two elements corresponds to most significant bits of the value which is simply additional information regarding the data, and the element(s) do(es) not apply the exception in a meaningful way (MPEP 2106.05(e)).
Accordingly, the additional element(s) do(es) not integrate the abstract idea into a practical application because the additional element(s) do(es) not impose any meaningful limits on practicing the abstract idea, and, therefore, the claim is directed to an abstract idea.
Step 2B Analysis: The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception. As discussed above with respect to the integration of the abstract idea into a practical application, the additional element(s) of:
additional information regarding the data do(es) not apply the exception in a meaningful way (MPEP 2106.05(e))
The additional element(s) do(es) not provide an inventive concept, and, therefore, the claim is not patent eligible.
Regarding claim 46, the claim is rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
Step 1 Analysis: Claim 46 is directed to computer-readable media, which is directed to an article of manufacture, one of the statutory categories.
Step 2A Prong One Analysis: The claim recites a(n) one or more non-transitory computer-readable media.
The limitation of wherein compressing the value comprises setting one of the two elements to zero, as drafted, is a process that, under its broadest reasonable interpretation, covers a mental process. The limitation is directed to observation, evaluation, judgment and opinion and is a process capable of being performed by a human mentally or using pen and paper.
If a claim limitation, under its broadest reasonable interpretation, covers performance of the limitation in the mind, then it falls within the "Mental Processes" grouping. Accordingly, the claim recites an abstract idea.
Step 2A Prong Two Analysis: With respect to the abstract idea, the judicial exception is not integrated into a practical application.
The claim recites wherein the one of the two elements corresponds to most significant bits of the value which is simply additional information regarding the data, and the element(s) do(es) not apply the exception in a meaningful way (MPEP 2106.05(e)).
Accordingly, the additional element(s) do(es) not integrate the abstract idea into a practical application because the additional element(s) do(es) not impose any meaningful limits on practicing the abstract idea, and, therefore, the claim is directed to an abstract idea.
Step 2B Analysis: The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception. As discussed above with respect to the integration of the abstract idea into a practical application, the additional element(s) of:
additional information regarding the data do(es) not apply the exception in a meaningful way (MPEP 2106.05(e))
The additional element(s) do(es) not provide an inventive concept, and, therefore, the claim is not patent eligible.
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 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) 24-26, 28-33, 35, 37-46 is/are rejected under 35 U.S.C. 103 as being unpatentable over Peluso et al. (Energy-Accuracy Scalable Deep Convolutional Neural Networks: A Pareto Analysis, hereinafter referred to as “Peluso”) in view of Kim et al. (Deep Convolutional Neural Network Accelerator Featuring Conditional Computing and Low External Memory Access, hereinafter referred to as "Kim").
Regarding claim 24 (Currently Amended), Peluso teaches an apparatus for deep learning (Peluso, section 3.2 – teaches hardware implementation; Peluso, section 3.3 – teaches an apparatus including hardware and software implementing the DL strategy; see also Peluso, Fig. 2), comprising:
a mixed precision circuit configured (Peluso, section 3.2 – teaches hardware implementation; see also Peluso, Fig. 2) to:
decompose an element in an input feature map of a deep learning operation into two input elements, the element in the input feature map having a first precision, the two input elements having a second precision that is lower than the first precision (Peluso, section 3.1 – teaches decomposing an input element if an input feature map of a convolutional neural network into two halfwords of most significant bits and least significant bits), and
decompose a weight of the deep learning operation into two weight elements, the weight associated with the element in the input feature map, the weight having the first precision, the two weight elements having the second precision (Peluso, section 3.1 – teaches decomposing a weight element of a kernel a convolutional neural network into two halfwords of most significant bits and least significant bits);
a multiply-accumulation circuit (Peluso, section 3.2 – teaches hardware implementation; see also Peluso, Fig. 2) configured to perform a computation using the two input elements and the two weight elements (Peluso, section 3.1 – teaches performing computations using the halfwords for the input and weight; see also Peluso, Fig. 1);
a sparsity circuit (Peluso, section 3.2 – teaches hardware implementation; see also Peluso, Fig. 2) configured to skip one or more computations of one or more zero-valued elements in the input feature map by the multiply-accumulation circuit (Peluso, section 3.2 – teaches a zero-skipping strategy the skips computation if one of the operands is zero).
However, Peluso does not explicitly teach a compression circuit configured to compress values computed by the multiply-accumulate circuit, wherein the compressed values are output activations of the deep learning operation and are to be processed in a next deep learning operation.
Kim teaches a compression circuit configured to compress values computed by the multiply-accumulate circuit, wherein the compressed values are output activations of the deep learning operation and are to be processed in a next deep learning operation (Kim, section II.B – teaches ReLU activation functions which compress negative output activations to zero; see also Kim, section III).
It would have been obvious to one of ordinary skill in the art before the filing date of the claimed invention to modify Peluso with the teachings of Kim in order to reduce redundant operations and increase sparsity in the field of accelerating neural network computations (Kim, Abstract – “This paper presents an ASIC accelerator for deep convolutional neural networks (DCNNs) featuring a novel conditional computing scheme that synergistically combines precision-cascading with zero-skipping. To reduce many redundant convolution operations that are followed by max-pooling operations, we propose precision-cascading, where the input features are divided into a number of low-precision groups and approximate convolutions with only the most significant bits (MSBs) are performed first. Based on this approximate computation, the full-precision convolution is performed only on the maximum pooling output that is found... Precision-cascading provides the added benefit of increased sparsity per low-precision group, which we exploit with zero-skipping to eliminate clock cycles as well as external memory access that involve zero inputs...”).
Regarding claim 25 (Currently Amended), Peluso in view of Kim teaches all of the limitations of the apparatus of claim 24 as noted above. Kim further teaches wherein the compression circuit is further configured to compress a plurality of elements in the input feature map, the plurality of elements comprising the element (Kim, section III – teaches loading non-zero input element values into the processing element [Not loading zero value inputs is a compression of the input]).
It would have been obvious to one of ordinary skill in the art before the filing date of the claimed invention to combine the teachings of Peluso and Kim in order to compress the input feature map to reduce redundant operations and increase sparsity (Kim, Abstract).
Regarding claim 26 (Currently Amended), Peluso in view of Kim teaches all of the limitations of the apparatus of claim 25 as noted above. Kim further teaches wherein the compression circuit is further configured to compress a plurality of weights of the deep learning operation, the plurality of weights comprising the weight (Kim, section III – teaches skipping kernel features of [weights] of zero-value inputs [Not loading weights for zero value inputs is a compression of the weight]).
It would have been obvious to one of ordinary skill in the art before the filing date of the claimed invention to combine the teachings of Peluso and Kim in order to compress weights to reduce redundant operations and increase sparsity (Kim, Abstract).
Regarding claim 28 (Currently Amended), Peluso in view of Kim teaches all of the limitations of the apparatus of claim 27 as noted above. Peluso further teaches wherein the mixed precision unit is further configured to:
decompose the output activation into two output elements having the second precision (Peluso, section 3.1 – teaches decomposing an input element if an input feature map of a convolutional neural network into two halfwords of most significant bits and least significant bits; Peluso, section 4.2 – teaches decomposing inputs for each layer [decomposing inputs for a subsequent layer would require decomposing the output of the preceding layer]).
It would have been obvious to one of ordinary skill in the art before the filing data of the claimed invention to combine the teachings of Peluso and Kim for the same reasons as disclosed in claim 24 above.
Regarding claim 29 (Currently Amended), Peluso in view of Kim teaches all of the limitations of the apparatus of claim 28 as noted above. Peluso further teaches wherein compression circuit is configured to compress the values by setting one of the two elements to zero (Peluso, section 3.1 – teaches setting elements to zero based on the desired precision).
It would have been obvious to one of ordinary skill in the art before the filing data of the claimed invention to combine the teachings of Peluso and Kim for the same reasons as disclosed in claim 28 above.
Regarding claim 30 (Currently Amended), Peluso in view of Kim teaches all of the limitations of the apparatus of claim 24 as noted above. Kim further teaches an additional circuit configured to compute the element in the input feature map based on an activation function (Kim, section II.B – teaches ReLU activation functions to generate input feature map elements for the next layer; see also Kim, section III).
It would have been obvious to one of ordinary skill in the art before the filing date of the claimed invention to combine the teachings of Peluso and Kim in order to compute input feature elements to reduce redundant operations and increase sparsity (Kim, Abstract).
Regarding claim 31 (Currently Amended), Peluso teaches a method for deep learning, comprising:
decomposing an element in an input feature map of a deep learning operation into two input elements, the element in the input feature map having a first precision, the two input elements having a second precision that is lower than the first precision (Peluso, section 3.1 – teaches decomposing an input element if an input feature map of a convolutional neural network into two halfwords of most significant bits and least significant bits), the input feature map further comprising one or more zero-valued elements (Peluso, section 3.2 – teaches a zero-skipping strategy the skips computation if one of the operands is zero);
decomposing a weight of the deep learning operation into two weight elements, the weight associated with the element in the input feature map, the weight having the first precision, the two weight elements having the second precision (Peluso, section 3.1 – teaches decomposing a weight element of a kernel a convolutional neural network into two halfwords of most significant bits and least significant bits);
performing a computation using the two input elements and the two weight elements to compute a value (Peluso, section 3.1 – teaches performing computations using the halfwords for the input and weight; see also Peluso, Fig. 1);
skipping one or more computations of the one or more zero-valued elements (Peluso, section 3.2 – teaches a zero-skipping strategy the skips computation if one of the operands is zero).
However, Peluso does not explicitly teach compressing the value, wherein the compressed value is an output activation of the deep learning operation and is to be processed in a next deep learning operation.
Kim teaches compressing the value, wherein the compressed value is an output activation of the deep learning operation and is to be processed in a next deep learning operation (Kim, section II.B – teaches ReLU activation functions which compress negative output activations to zero; see also Kim, section III).
It would have been obvious to one of ordinary skill in the art before the filing date of the claimed invention to modify Peluso with the teachings of Kim in order to reduce redundant operations and increase sparsity in the field of accelerating neural network computations (Kim, Abstract – “This paper presents an ASIC accelerator for deep convolutional neural networks (DCNNs) featuring a novel conditional computing scheme that synergistically combines precision-cascading with zero-skipping. To reduce many redundant convolution operations that are followed by max-pooling operations, we propose precision-cascading, where the input features are divided into a number of low-precision groups and approximate convolutions with only the most significant bits (MSBs) are performed first. Based on this approximate computation, the full-precision convolution is performed only on the maximum pooling output that is found... Precision-cascading provides the added benefit of increased sparsity per low-precision group, which we exploit with zero-skipping to eliminate clock cycles as well as external memory access that involve zero inputs...”).
Regarding claim 32 (Previously Presented), Peluso in view of Kim teaches all of the limitations of the method of claim 31 as noted above. Kim further teaches
compressing the input feature map (Kim, section III – teaches loading non-zero input element values into the processing element [Not loading zero value inputs is a compression of the input]); and
storing a compressed version of the input feature map in a memory (Kim, section III teaches loading non-zero input values [compressed input] into the processing element array).
It would have been obvious to one of ordinary skill in the art before the filing date of the claimed invention to combine the teachings of Peluso and Kim in order to compress input feature elements to reduce redundant operations and increase sparsity (Kim, Abstract).
Regarding claim 33 (Previously Presented), the rejection of claim 31 is incorporated herein. Further, the limitations in this claim are taught by Peluso in view of Kim for the reasons set forth in the rejection of claim 26.
Regarding claim 35 (Currently Amended), the rejection of claim 34 is incorporated herein. Further, the limitations in this claim are taught by Peluso in view of Kim for the reasons set forth in the rejection of claim 28.
Regarding claim 37 (Previously Presented), the rejection of claim 31 is incorporated herein. Further, the limitations in this claim are taught by Peluso in view of Kim for the reasons set forth in the rejection of claim 30.
Regarding claim 38 (Currently Amended), it is the computer-readable media embodiment of claim 31 with similar limitations to claim 31 and is rejected using the same reasoning found in claim 31. Peluso further teaches one or more non-transitory computer-readable media storing instructions executable to perform operations (Peluso, section 3.2 – teaches hardware implementation; Peluso, section 3.3 – teaches an apparatus including hardware and software implementing the DL strategy; see also Peluso, Fig. 2) …
It would have been obvious to one of ordinary skill in the art before the filing data of the claimed invention to combine the teachings of Peluso and Kim for the same reasons as disclosed in claim 31 above.
Regarding claim 39 (Previously Presented), the rejection of claim 38 is incorporated herein. Further, the limitations in this claim are taught by Peluso in view of Kim for the reasons set forth in the rejection of claim 32.
Regarding claim 40 (Previously Presented), the rejection of claim 38 is incorporated herein. Further, the limitations in this claim are taught by Peluso in view of Kim for the reasons set forth in the rejection of claim 33.
Regarding claim 41 (Currently Amended), Peluso in view of Kim teaches all of the limitations of the method of claim 31 as noted above. Peluso further teaches generating an output activation of the deep learning operation from the computation (Peluso, sections 3.1-3.2 – teaches the MAC outputting the dot product of the input and weight elements; see also Peluso, Algorithm 1).
It would have been obvious to one of ordinary skill in the art before the filing data of the claimed invention to combine the teachings of Peluso and Kim for the same reasons as disclosed in claim 31 above.
Regarding claim 42 (Currently Amended), the rejection of claim 41 is incorporated herein. Further, the limitations in this claim are taught by Peluso in view of Kim for the reasons set forth in the rejection of claim 35.
Regarding claim 43 (Previously Presented), the rejection of claim 38 is incorporated herein. Further, the limitations in this claim are taught by Peluso in view of Kim for the reasons set forth in the rejection of claim 37.
Regarding claim 44 (New), Peluso in view of Kim teaches all of the limitations of the apparatus of claim 29 as noted above. Peluso further teaches wherein the one of the two elements corresponds to most significant bits of the value (Peluso, section 3.1 – teaches decomposing an input element if an input feature map of a convolutional neural network into two halfwords of most significant bits and least significant bits; Peluso, section 4.2 – teaches decomposing inputs for each layer [decomposing inputs for a subsequent layer would require decomposing the output of the preceding layer]).
It would have been obvious to one of ordinary skill in the art before the filing data of the claimed invention to combine the teachings of Peluso and Kim for the same reasons as disclosed in claim 29 above.
Regarding claim 45 (New), Peluso in view of Kim teaches all of the limitations of the method of claim 35 as noted above. Peluso further teaches wherein compressing the value comprises setting one of the two elements to zero (Peluso, section 3.1 – teaches setting elements to zero based on the desired precision), and wherein the one of the two elements corresponds to most significant bits of the value (Peluso, section 3.1 – teaches decomposing an input element if an input feature map of a convolutional neural network into two halfwords of most significant bits and least significant bits; Peluso, section 4.2 – teaches decomposing inputs for each layer [decomposing inputs for a subsequent layer would require decomposing the output of the preceding layer]).
It would have been obvious to one of ordinary skill in the art before the filing data of the claimed invention to combine the teachings of Peluso and Kim for the same reasons as disclosed in claim 35 above.
Regarding claim 46 (New), Peluso in view of Kim teaches all of the limitations of the computer-readable media of claim 42 as noted above. Peluso further teaches wherein compressing the value comprises setting one of the two elements to zero (Peluso, section 3.1 – teaches setting elements to zero based on the desired precision), and wherein the one of the two elements corresponds to most significant bits of the value (Peluso, section 3.1 – teaches decomposing an input element if an input feature map of a convolutional neural network into two halfwords of most significant bits and least significant bits; Peluso, section 4.2 – teaches decomposing inputs for each layer [decomposing inputs for a subsequent layer would require decomposing the output of the preceding layer]).
It would have been obvious to one of ordinary skill in the art before the filing data of the claimed invention to combine the teachings of Peluso and Kim for the same reasons as disclosed in claim 42 above.
Conclusion
Applicant's amendment necessitated the new ground(s) of rejection presented in this Office action. Accordingly, THIS ACTION IS MADE FINAL. See MPEP § 706.07(a). Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any nonprovisional extension fee (37 CFR 1.17(a)) pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the mailing date of this final action.
Any inquiry concerning this communication or earlier communication from the examiner should be directed to MARSHALL WERNER whose telephone number is (469) 295-9143. The examiner can normally be reached on Monday – Thursday 7:30 AM – 4:30 PM ET.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Kamran Afshar, can be reached at (571) 272-7796. The fax 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.
/MARSHALL L WERNER/ Primary Examiner, Art Unit 2125
Prosecution Insights