# US 2021/0279575 A1 · Information processing apparatus, information processing method, and storage medium > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 2021/0279575 A1 - Document ID: US20210279575A1 - Status: Published application - Kind: A1 - Application number: 17/189,571 - Filed: 2021-03-02 - Published: 2021-09-09 - Earliest claimed: 2020-03-05 - Assignees: Canon Kabushiki Kaisha - Applicants: Canon Kabushiki Kaisha - Inventors: Takayuki Komatsu - Web page: https://patentgrep.com/patent/US20210279575A1 ## CPC - G06N 3/08: Learning methods - G06F 18/217: {Validation; Performance evaluation; Active pattern learning techniques} - G06F 18/285: {Selection of pattern recognition techniques, e.g. of classifiers in a multi-classifier system} - G06N 3/04: Architecture, e.g. interconnection topology - G06N 3/0495: Quantised networks; Sparse networks; Compressed networks - G06N 3/0499: Feedforward networks - G06N 3/09: Supervised learning - G06N 5/01: Dynamic search techniques; Heuristics; Dynamic trees; Branch-and-bound - G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS - G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N) ## Abstract An information processing apparatus designates one or more constraints for constraining a configuration of a neural network, which include information for specifying the configuration of the neural network and the maximum number of computations in the neural network, and executes a computation of a neural network configured based on the designated constraints. When a neural network is trained, the information processing apparatus trains respective neural networks under the one or more designated constraints, and selects, for estimation, a predetermined learned model out of the learned models trained under the one or more designated constraints. ## Claims ### Claim 1 An information processing apparatus comprising: one or more processors; and a memory storing instructions which, when the instructions are executed by the one or more processors, cause the information processing apparatus to function as: a designation unit configured to designate one or more constraints for constraining a configuration of a neural network, the constraints including information for specifying the configuration of the neural network and the maximum number of computations in the neural network; and a processing unit configured to execute a computation of a neural network configured based on the designated constraints, wherein, when a neural network is trained, the processing unit trains respective neural networks under the one or more designated constraints, and selects, for estimation, a predetermined learned model out of the learned models trained under the one or more designated constraints. ### Claim 2 (depends on 1) The information processing apparatus according to claim 1, wherein the maximum number of computations indicates the number of product-sum computations required when a computation of the neural network is executed. ### Claim 3 (depends on 2) The information processing apparatus according to claim 2, wherein a product-sum computation of the neural network includes a computation of adding inputs to which the same weight coefficient of the neural network is to be applied, to each other in advance, and multiplying the resultant by the weight coefficient. ### Claim 4 (depends on 1) The information processing apparatus according to claim 1, wherein the processing unit selects, as the predetermined learned model, a learned model for which an evaluation value for evaluating a degree of optimization of the learned model is lower than a predetermined evaluation value, from among the respective learned models trained under the one or more designated constraints. ### Claim 5 (depends on 1) The information processing apparatus according to claim 1, wherein the processing unit selects, as the predetermined learned model, a learned model for which the number of times of computation is smaller than another learned model from among the respective learned models trained under the one or more designated constraints. ### Claim 6 (depends on 1) The information processing apparatus according to claim 1, wherein the information for specifying the configuration of the neural network includes at least some of the number of nodes of an input layer of the neural network, the number of nodes of an output layer of the neural network, the number of layers that constitute the neural network, and the number of nodes of each hidden layer of the neural network. ### Claim 7 (depends on 1) The information processing apparatus according to claim 1, wherein weight coefficient of the neural network includes a quantized weight coefficient. ### Claim 8 An information processing method comprising: designating one or more constraints for constraining a configuration of a neural network, the constraints including information for specifying the configuration of the neural network and the maximum number of computations in the neural network; and processing for executing a computation of a neural network configured based on the one or more designated constraints, wherein, the processing comprises, when a neural network is trained, training respective neural networks under the one or more designated constraints, and selecting, for estimation, a predetermined learned model out of the learned models trained under the one or more designated constraints. ### Claim 9 A non-transitory computer-readable storage medium comprising instructions for performing an information processing method, the information processing method comprising: designating one or more constraints for constraining a configuration of a neural network, the constraints including information for specifying the configuration of the neural network and the maximum number of computations in the neural network; and processing for executing a computation of a neural network configured based on the one or more designated constraints, wherein, processing comprises, when a neural network is trained, training respective neural networks under the one or more designated constraints, and selecting, for estimation, a predetermined learned model out of the learned models trained under the one or more designated constraints.