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
1. This Office Action is in response to the application filed on 10/21/2024. Claims 1-20 are pending in this application. Claims 1, 8 and 15 are independent claims.
Claim Rejections - 35 USC § 101
2. 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.
3. Claims 1-20 are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more. The independent claims 1, 8 and 15 are corresponding to one of four statutory categories including method, system, and method respectively under step 1. The claims 1, 8 and 15 similarly recite “a method of making an executable neural network, the method comprising operations for: partitioning at least a portion of an input tensor of a convolution in the executable neural network into micro-tiles; performing a plurality of iterations of the convolution using the micro-tiles, wherein performing the plurality of iterations comprises performing, by convolution engines, multiply-accumulate (MAC) operations on the micro-tiles in parallel, different micro-tiles processed by different ones of the convolution engines; and tracking which iterations of the plurality of iterations have been completed by the convolution engines”.
The limitation of the claims 1, 8 and 15 of “partitioning at least a portion of an input tensor of a convolution in the executable neural network into micro-tiles;” as drafted, is a mental process that, under its broadest reasonable interpretation, covers a mental process but for the recitation of generic computer components. For example, but for the “partitioning (dividing)” in the context of this claim encompasses the user may partition at least a portion of an input tensor of a convolution in the neural network into micro-tiles like breaking a large multi-dimensional array/vector into smaller, manageable sub-blocks with a pen and paper or in a human mind. If a claim limitation, under its broadest reasonable interpretation, covers performance of the limitation in the mind but for the recitation of generic computer components, then it falls within the “Mental Processes” grouping of abstract ideas. Accordingly, the claim recites an abstract idea under Step 2A Prong 1.
The limitation of the claims 1, 8 and 15 of “performing a plurality of iterations of the convolution using the micro-tiles” as drafted, is a mathematical operation that, under its broadest reasonable interpretation, covers a mathematical operation but for the recitation of generic computer components. For example, but for the “iterating (repeating) convolutions (calculating)” in the context of this claim encompasses the user may perform a plurality of iterations of the convolution using the micro-tiles with a pen and paper or in a human mind. If a claim limitation, under its broadest reasonable interpretation, covers performance of the limitation in the mind but for the recitation of generic computer components, then it falls within the “Mathematical Operations” grouping of abstract ideas. Accordingly, the claim recites an abstract idea under Step 2A Prong 1.
The limitation of the claims 1, 8 and 15 of “wherein performing the plurality of iterations comprises performing, by convolution engines, multiply-accumulate (MAC) operations on the micro-tiles in parallel, different micro-tiles processed by different ones of the convolution engines;” as drafted, is a mathematical operation that, under its broadest reasonable interpretation, covers a mathematical operation but for the recitation of generic computer components. For example, but for the “iteratively calculating multiply-accumulate” in the context of this claim encompasses the user may perform the plurality of iterations comprises performing, by convolution engines, multiply-accumulate (MAC) operations on the micro-tiles in parallel, different micro-tiles processed by different ones of the convolution engines with a pen and paper or in a human mind. If a claim limitation, under its broadest reasonable interpretation, covers performance of the limitation in the mind but for the recitation of generic computer components, then it falls within the “Mathematical Operations” grouping of abstract ideas. Accordingly, the claim recites an abstract idea under Step 2A Prong 1.
The limitation of the claims 1, 8 and 15 of “tracking which iterations of the plurality of iterations have been completed by the convolution engines” as drafted, is a mental process that, under its broadest reasonable interpretation, covers a mental process but for the recitation of generic computer components. For example, but for the “tracking” in the context of this claim encompasses the user may track which iterations of the plurality of iterations have been completed by the convolution engines with a pen and paper or in a human mind. If a claim limitation, under its broadest reasonable interpretation, covers performance of the limitation in the mind but for the recitation of generic computer components, then it falls within the “Mental Processes” grouping of abstract ideas. Accordingly, the claim recites an abstract idea under Step 2A Prong 1.
The limitation of the claims 2, 9 and 16 of “performing a plurality of additional iterations of the convolution on another portion of the input tensor” as drafted, is a mathematical operation that, under its broadest reasonable interpretation, covers a mathematical operation but for the recitation of generic computer components. For example, but for the “iterating (repeating) convolutions” in the context of this claim encompasses the user may perform a plurality of additional iterations of the convolution on another portion of the input tensor with a pen and paper or in a human mind. If a claim limitation, under its broadest reasonable interpretation, covers performance of the limitation in the mind but for the recitation of generic computer components, then it falls within the “Mathematical Operations” grouping of abstract ideas. Accordingly, the claim recites an abstract idea under Step 2A Prong 1.
The limitation of the claims 3, 10 and 17 of “the plurality of additional iterations is performed in parallel with the plurality of iterations” as drafted, is a mental process that, under its broadest reasonable interpretation, covers a mental process but for the recitation of generic computer components. For example, but for the “iterating (repeating)” in the context of this claim encompasses the user may perform the plurality of additional iterations in parallel with the plurality of iterations with a pen and paper or in a human mind. If a claim limitation, under its broadest reasonable interpretation, covers performance of the limitation in the mind but for the recitation of generic computer components, then it falls within the “Mental Processes” grouping of abstract ideas. Accordingly, the claim recites an abstract idea under Step 2A Prong 1.
The limitation of the claims 4, 11 and 18 of “tracking whether the plurality of iterations or the plurality of additional iterations has been completed” as drafted, is a mental process that, under its broadest reasonable interpretation, covers a mental process but for the recitation of generic computer components. For example, but for the “tracking” in the context of this claim encompasses the user may track whether the plurality of iterations or the plurality of additional iterations has been completed with a pen and paper or in a human mind. If a claim limitation, under its broadest reasonable interpretation, covers performance of the limitation in the mind but for the recitation of generic computer components, then it falls within the “Mental Processes” grouping of abstract ideas. Accordingly, the claim recites an abstract idea under Step 2A Prong 1.
The limitation of the claims 5, 12 and 19 of “partitioning the additional tile into additional micro-tiles” as drafted, is a mental process that, under its broadest reasonable interpretation, covers a mental process but for the recitation of generic computer components. For example, but for the “partitioning” in the context of this claim encompasses the user may partition the additional tile into additional micro-tiles with a pen and paper or in a human mind. If a claim limitation, under its broadest reasonable interpretation, covers performance of the limitation in the mind but for the recitation of generic computer components, then it falls within the “Mental Processes” grouping of abstract ideas. Accordingly, the claim recites an abstract idea under Step 2A Prong 1.
The limitation of the claims 5, 12 and 19 of “performing, by additional convolution engines, MAC operations on the additional micro-tiles in parallel, different ones of the additional micro-tiles processed by different ones of the additional convolution engines” as drafted, is a mathematical operation that, under its broadest reasonable interpretation, covers a mathematical operation but for the recitation of generic computer components. For example, but for the “performing MAC operations (calculating)” in the context of this claim encompasses the user may perform, by additional convolution engines, MAC operations on the additional micro-tiles in parallel, different ones of the additional micro-tiles processed by different ones of the additional convolution engines with a pen and paper or in a human mind. If a claim limitation, under its broadest reasonable interpretation, covers performance of the limitation in the mind but for the recitation of generic computer components, then it falls within the “Mathematical Operations” grouping of abstract ideas. Accordingly, the claim recites an abstract idea under Step 2A Prong 1.
The limitation of the claims 6, 13 and 20 of “tracking which iterations of the plurality of additional iterations have been completed by the additional convolution engines” as drafted, is a mental process that, under its broadest reasonable interpretation, covers a mental process but for the recitation of generic computer components. For example, but for the “tracking” in the context of this claim encompasses the user may track which iterations of the plurality of additional iterations have been completed by the additional convolution engines with a pen and paper or in a human mind. If a claim limitation, under its broadest reasonable interpretation, covers performance of the limitation in the mind but for the recitation of generic computer components, then it falls within the “Mental Processes” grouping of abstract ideas. Accordingly, the claim recites an abstract idea under Step 2A Prong 1.
This judicial exception is not integrated into a practical application. In particular, the claims 7 and 14 recite additional elements such as “the input tensor includes an image input into the executable neural network”.
Examiner would like to point out that with the broad reasonable interpretation, this element
amounts to field of use under MPEP § 2106.05(h): Field of Use and Technological Environment, which
does not impose any meaningful limits on practicing the mental process. Accordingly, this additional
element does not integrate the abstract idea into a practical application because it does not impose any
meaningful limits on practicing the abstract idea. The claim is directed to an abstract idea under Step 2A
Prong 2 and 2B.
Dependent claims 2-7, 9-14 and 16-20 are also similar rejected under same rationale as cited above wherein these claims do not include additional elements that are sufficient to amount to significantly more than the judicial exception. These claims are merely further elaborate the mental process itself or providing additional definition of process which does not impose any meaningful limits on practicing the abstract idea. Claims 2-7, 9-14 and 16-20 are also rejected for incorporating the deficiency of their independent claims 1, 8 and 15 respectively.
Reasons for Allowance
4. The following is an examiner’s statement of reasons for allowance: the prior-art, Kuo (US PGPub 20190220742), in view of Ross (US Patent 11537687), and further in view of Nagy (US PGPub 20210241082) failed to disclose: a method of making an executable neural network, the method comprising operations for: partitioning at least a portion of an input tensor of a convolution in the executable neural network into micro-tiles; performing a plurality of iterations of the convolution using the micro-tiles, wherein performing the plurality of iterations comprises performing, by convolution engines, multiply-accumulate (MAC) operations on the micro-tiles in parallel, different micro-tiles processed by different ones of the convolution engines; and tracking which iterations of the plurality of iterations have been completed by the convolution engines, as recited by the independent claim 1.
Regarding Claim 1, the closest prior-art found, Kuo, Ross and Nagy discloses of a method of making an executable neural network, the method comprising operations for: partitioning at least a portion of an input tensor of a convolution in the executable neural network into micro-tiles; performing an iteration of the convolution on the tile, wherein performing the iteration of the convolution comprises: performing, by convolution engines, multiply-accumulate (MAC) operations on the micro-tiles; and tracking which iterations of the plurality of iterations have been completed by the convolution engines.
Individually, Kuo teaches that the input tiles may overlap with each other, and each tile is divided into equal-sized, non-overlapping blocks. A block (e.g., block 211) is a basic unit of computation. For example, an engine (e.g., the convolution engine 111) may include an array of multiply-and-accumulate (MAC) circuits, and the size of a block may be equal to the size of the MAC array. Thus, [in par 19 and 24, the convolution engine to perform convolution is processing the multiple input overlapped tiles] operations on a block can be performed in parallel within an [convolution] engine. FIGS. 3A and 3B illustrate examples of overlapped input tiles in an input feature map 310 according to some embodiments. The input feature map 310 may be an input to the convolution engine 111 in FIG. 1.
Ross teaches that the control pattern 136 is a matrix that indicates to the multiply-add unit 140 which portions of the flattened input stream 124 to select from in order to generate the selected values that are multiplied (using the dot product) with the vector of the expanded kernel 128 to generate each output value. In a naïve implementation of convolution, for each stride of the kernel 110 at a position on the input tensor 102, the convolution operation is performed by summing the values adjacent to the position representing the current focus of the kernel along with the value at the focus point itself, as weighted by the corresponding values in the kernel (which are at the same positions).
Nagy teaches that as the control module 411 schedules a command to execute in one of the plurality of convolution modules 409, the respective input control unit 417 and coefficient control unit 427 of the convolution module 409 retrieve input data of at least a part of the input planes and their weight tensors, respectively, from the memory 405. The input planes may have a size of IW×IH and may be divided into a plurality of tiles and the convolution module 409 may retrieve one or more of the tiles. The memory of the MAC unit is different from conventional MACs that use registers to buffer results for subsequent accumulation.
However, the prior art, Kuo, Ross and Nagy failed to disclose the following subject matter such as “performing the plurality of iterations comprises performing, by convolution engines, multiply-accumulate (MAC) operations on the micro-tiles in parallel, different micro-tiles processed by different ones of the convolution engines”.
Claim 8 is a product system claim, similar to the claim 1, and Claim 15 is the system claim, similar to the claim 1. Therefore, claims 1-20 contain allowable subject matter over the prior art.
5. Any comments considered necessary by applicant must be submitted no later than the payment of the issue fee and, to avoid processing delays, should preferably accompany the issue fee. Such submissions should be clearly labeled “Comments on Statement of Reasons for Allowance.”
Pertinent Prior Art
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
Xie (US PGPub 20180157969): Xie teaches of a convolution and pooling unit for performing a convolution and pooling operation for a first iteration number of times on input data in accordance with convolution parameter information to finally obtain an input vector of a sparse neural network, wherein each input data is divided into a plurality of sub-blocks, and the convolution and pooling unit performs the convolution and pooling operation on the plurality of sub-blocks in parallel; a full connection unit for performing a full connection calculation for a second iteration number of times on the input vector in accordance with weight matrix position information of a full connection layer to finally obtain a calculation result of the sparse convolutional neural network, wherein each input vector is divided into a plurality of sub-blocks, and the full connection unit performs a full connection operation on the plurality of sub-blocks in parallel; and a control unit for determining and sending the convolution parameter information and the weight matrix position information of the full connection layer to the convolution and pooling unit and the full connection unit respectively, and controlling reading of the input vectors on respective iterative levels in the units above and their state machines.
Chao (US PGPub 20200065251): Chao teaches that the processing unit includes a multiply-accumulate unit and a processing loop controller connected to the multiply-accumulate unit. The multiply-accumulate unit is configured by the processing loop controller to perform a convolutional layer operation with the input feature map tile to produce the output feature map tile according to a memory-adaptive processing technique.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to JAE UK JEON whose telephone number is (571)270-3649. The examiner can normally be reached 10am-6pm. Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Chat Do can be reached at 571-272-3721. 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.
/JAE U JEON/Primary Examiner, Art Unit 2193
Prosecution Insights