Filters








2 Hits in 3.8 sec

MF-Net: Compute-In-Memory SRAM for Multibit Precision Inference using Memory-immersed Data Conversion and Multiplication-free Operators [article]

Shamma Nasrin, Diaa Badawi, Ahmet Enis Cetin, Wilfred Gomes, Amit Ranjan Trivedi
2021 arXiv   pre-print
We propose a co-design approach for compute-in-memory inference for deep neural networks (DNN).  ...  We use multiplication-free function approximators based on ell_1 norm along with a co-adapted processing array and compute flow.  ...  Additionally, multibit precision DNN inference is complex using compute-in-memory.  ... 
arXiv:2102.00035v1 fatcat:hveresaqmngbfi5tqopniuwn5q

Table of contents

2021 IEEE Transactions on Circuits and Systems Part 1: Regular Papers  
Inference Using Memory-Immersed Data Conversion and Multiplication-Free Operators .....................................  ...  Liang 1906 (Contents Continued on Back Cover) (Contents Continued from Front Cover) Digital Circuits and Systems and VLSI A 7.8-13.6 pJ/b MF-Net: Compute-In-Memory SRAM for Multibit Precision  ... 
doi:10.1109/tcsi.2021.3072264 fatcat:uhdta35gybe6rddwiwqfntuu2e