INFORMATION PROCESSING APPARATUS, INFORMATION PROCESSING METHOD, AND STORAGE MEDIUM: abstract simplified (18187577)

• This abstract for appeared for patent application number 18187577 Titled 'INFORMATION PROCESSING APPARATUS, INFORMATION PROCESSING METHOD, AND STORAGE MEDIUM'

Simplified Explanation

The abstract describes a method for optimizing the structure of a neural network to achieve accurate results without increasing processing requirements. The method involves generating multiple candidates for a specific part of the neural network, inputting training data to the network using these candidates, and obtaining an inference result. The network's loss is calculated based on the number of candidates selected and the inference result, and the weight coefficients for the candidates are updated accordingly. Finally, candidates are selected based on the updated weight coefficients.

Original Abstract Submitted

The present disclosure makes it possible to learn a neural network architecture for achieving a sufficient inference accuracy while preventing an increase in the amount of processing. An information processing apparatus configured to learn an architecture for optimizing a structure of a neural network generates a plurality of candidates for an edge of the neural network, inputs learning data to the neural network with weight coefficients set to these candidates for the edge, and obtains an inference result. The information processing apparatus calculates a loss of the neural network based on a specified candidate number which is the number of candidates to be selected from the plurality of candidates and on the inference result, and then updates the weight coefficients for the plurality of candidates based on the loss. The information processing apparatus then selects candidates from the plurality of candidates based on the updated weight coefficients.