Invention Application
- Patent Title: NEURAL NETWORK PROCESSOR USING DYADIC WEIGHT MATRIX AND OPERATION METHOD THEREOF
-
Application No.: US16675709Application Date: 2019-11-06
-
Publication No.: US20200167637A1Publication Date: 2020-05-28
- Inventor: Junseok PARK , Inyup KANG
- Applicant: SAMSUNG ELECTRONICS CO., LTD.
- Priority: com.zzzhc.datahub.patent.etl.us.BibliographicData$PriorityClaim@5d1a80f5 com.zzzhc.datahub.patent.etl.us.BibliographicData$PriorityClaim@5a24a48c
- Main IPC: G06N3/063
- IPC: G06N3/063 ; G06N3/08 ; G06F17/16 ; G06F7/44

Abstract:
An neural network (NN) processor includes an input feature map buffer configured to store an input feature matrix, a weight buffer configured to store a weight matrix trained in a form of a, a transform circuit configured to perform a Walsh-Hadamard transform on an input feature vector obtained from the input feature matrix and a weight vector included in the weight matrix to output a transformed input feature vector and a transformed weight vector, and an arithmetic circuit configured to perform an element-wise multiplication (EWM) on the transformed input feature vector and the transformed weight vector.
Public/Granted literature
- US11562046B2 Neural network processor using dyadic weight matrix and operation method thereof Public/Granted day:2023-01-24
Information query