公开(公告)号：US20220027710A1

公开(公告)日：2022-01-27

申请号：US17376302

申请日：2021-07-15

Applicant: Samsung Electronics Co., Ltd.

Inventor： Fengtao XIE ,  Fangfang DU ,  Fang LIU ,  Liang LI ,  Pengfei ZHAO ,  Ke LU

IPC: G06N3/04 ,  G06F17/18 ,  G06N3/08

Abstract: A method and apparatus for determining a neural network architecture of a processor are provided. The method of determining a target neural network architecture, the method comprising obtaining a first neural network architecture, searching for the first neural network architecture based on a loss function, in response to a first search end condition not being satisfied, and determining a target neural network architecture used in a processor, based on a result of the searching, wherein the loss function is based on a processor computation cost.

公开(公告)号：US20230093407A1

公开(公告)日：2023-03-23

申请号：US17946218

申请日：2022-09-16

Applicant: Samsung Electronics Co., Ltd.

Inventor： Fangfang DU ,  Fang LIU ,  Liang LI ,  Pengfei ZHAO ,  Fengtao XIE

IPC: G06N3/063 ,  G06N3/08

Abstract: A method and apparatus with neural network optimization are provided. A method is performed by a device storing a target network block and processing hardware that performs optimizing for the target network block, the method includes generating, by the processing hardware, an extended network block of the target network block by increasing, a number of channels of a target operation branch in the target network block to a determined number of channels, wherein the target network block includes operation branches that include the target operation branch, and wherein each operation branch includes at least one respective channel, determining importance measures of the respective operation branches, including the target operation branch with the increased number of channels, in the extended network block, and clipping a channel of the target operation branch in the extended network block, the clipping is performed according to the importance measures of the respective operation branches.