Invention Grant
- Patent Title: Compression of deep neural networks with proper use of mask
-
Application No.: US15390660Application Date: 2016-12-26
-
Publication No.: US10832123B2Publication Date: 2020-11-10
- Inventor: Shijie Sun , Song Han , Xin Li , Yi Shan
- Applicant: BEIJING DEEPHI INTELLIGENCE TECHNOLOGY CO., LTD.
- Applicant Address: CN Beijing
- Assignee: XILINX TECHNOLOGY BEIJING LIMITED
- Current Assignee: XILINX TECHNOLOGY BEIJING LIMITED
- Current Assignee Address: CN Beijing
- Agency: IPro, PLLC
- Priority: com.zzzhc.datahub.patent.etl.us.BibliographicData$PriorityClaim@6eb89ecf com.zzzhc.datahub.patent.etl.us.BibliographicData$PriorityClaim@563938c7 com.zzzhc.datahub.patent.etl.us.BibliographicData$PriorityClaim@218f7126 com.zzzhc.datahub.patent.etl.us.BibliographicData$PriorityClaim@44a9a930
- Main IPC: G06N3/04
- IPC: G06N3/04 ; G06N3/08
Abstract:
The present invention relates to artificial neural networks, for example, deep neural networks. In particular, the present invention relates to a compression method for deep neural networks with proper use of mask and the device thereof. More specifically, the present invention relates to how to compress dense neural networks into sparse neural networks while maintaining or even improving the accuracy of the neural networks after compression.
Public/Granted literature
- US20180046915A1 COMPRESSION OF DEEP NEURAL NETWORKS WITH PROPER USE OF MASK Public/Granted day:2018-02-15
Information query
