Invention Grant
- Patent Title: Method to balance sparsity for efficient inference of deep neural networks
-
Application No.: US16186470Application Date: 2018-11-09
-
Publication No.: US11449756B2Publication Date: 2022-09-20
- Inventor: Weiran Deng
- Applicant: Samsung Electronics Co., Ltd.
- Applicant Address: KR Suwon-si
- Assignee: Samsung Electronics Co., Ltd.
- Current Assignee: Samsung Electronics Co., Ltd.
- Current Assignee Address: KR Suwon-si
- Agency: Renaissance IP Law Group LLP
- Main IPC: G06N3/08
- IPC: G06N3/08

Abstract:
A system and method that provides balanced pruning of weights of a deep neural network (DNN) in which weights of the DNN are partitioned into a plurality of groups, a count of a number of non-zero weights is determined in each group, a variance of the count of weights in each group is determined, a loss function of the DNN is minimized using Lagrange multipliers with a constraint that the variance of the count of weights in each group is equal to 0, and the weights and the Lagrange multipliers are retrained by back-propagation.
Public/Granted literature
- US20200097830A1 METHOD TO BALANCE SPARSITY FOR EFFICIENT INFERENCE OF DEEP NEURAL NETWORKS Public/Granted day:2020-03-26
Information query