TY - GEN
T1 - Efficient Multi-task Adaption for Crossbar-based In-Memory Computing
AU - Zhang, Fan
AU - Yang, Li
AU - Fan, Deliang
N1 - Funding Information:
This work is supported in part by the National Science Foundation under Grant No.2003749 and No. 2144751.
Publisher Copyright:
© 2022 IEEE.
PY - 2022
Y1 - 2022
N2 - ReRAM crossbar Non-volatile memory (NVM) based In-Memory Computing (IMC) has been widely investigated as a highly parallel, fast, and energy-efficient computing platform for Deep Neural Networks (DNNs), especially for one specific task inference. However, due to the intrinsic high energy consumption of weight re-programming and the relatively low endurance issue, adapting the ReRAM crossbar-based IMC hardware for continual learning or multi-task learning has not been well explored. In this paper, we discuss a crossbar-aware learning method with a 2-tier masking technique that could enable efficient and fast new task adaption for a deployed DNN model with minor hardware overhead.
AB - ReRAM crossbar Non-volatile memory (NVM) based In-Memory Computing (IMC) has been widely investigated as a highly parallel, fast, and energy-efficient computing platform for Deep Neural Networks (DNNs), especially for one specific task inference. However, due to the intrinsic high energy consumption of weight re-programming and the relatively low endurance issue, adapting the ReRAM crossbar-based IMC hardware for continual learning or multi-task learning has not been well explored. In this paper, we discuss a crossbar-aware learning method with a 2-tier masking technique that could enable efficient and fast new task adaption for a deployed DNN model with minor hardware overhead.
KW - Deep Neural Networks
KW - In-Memory Computing
KW - Multi-task Continual Learning
UR - http://www.scopus.com/inward/record.url?scp=85150188855&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85150188855&partnerID=8YFLogxK
U2 - 10.1109/IEEECONF56349.2022.10052040
DO - 10.1109/IEEECONF56349.2022.10052040
M3 - Conference contribution
AN - SCOPUS:85150188855
T3 - Conference Record - Asilomar Conference on Signals, Systems and Computers
SP - 328
EP - 333
BT - 56th Asilomar Conference on Signals, Systems and Computers, ACSSC 2022
A2 - Matthews, Michael B.
PB - IEEE Computer Society
T2 - 56th Asilomar Conference on Signals, Systems and Computers, ACSSC 2022
Y2 - 31 October 2022 through 2 November 2022
ER -