NeuroScrub+: Mitigating Retention Faults Using Flexible Approximate Scrubbing in Neuromorphic Fabric Based on Resistive Memories

Neuromorphic computation-in-memory fabric based on emerging non-volatile memories (NVM) is considered an attractive option to accelerate neural networks (NNs) in hardware as they provide high-performance, low-power, and reduced data movement. Although NVMs offer many benefits, they are susceptible t...

Full description

Saved in:
Bibliographic Details
Published inIEEE transactions on computer-aided design of integrated circuits and systems Vol. 42; no. 5; p. 1
Main Authors Ahmed, Soyed Tuhin, Hefenbrock, Michael, Munch, Christopher, Tahoori, Mehdi B.
Format Journal Article
LanguageEnglish
Published New York IEEE 01.05.2023
The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Subjects
Artificial neural networks
Circuit faults
Computation-in-memory
Faults
Hardware
Inference
Magnetic tunneling
MTJ
Multilayer perceptrons
Neuromorphic
Neuromorphic computing
Nonvolatile memory
Numerical stability
NVM
Resistance
Retention
retention faults
Thermal stability
Washing
Online AccessGet full text

Cover

More Information
Summary:Neuromorphic computation-in-memory fabric based on emerging non-volatile memories (NVM) is considered an attractive option to accelerate neural networks (NNs) in hardware as they provide high-performance, low-power, and reduced data movement. Although NVMs offer many benefits, they are susceptible to data retention faults, where previously stored data is not retained after a certain amount of time due to external influence. These faults are more likely to happen unidirectional and severely impact the inference accuracy of the hardware implementation of NNs since the synaptic weights stored in the NVMs are subject to retention faults. In this work, we propose an approximate scrubbing technique for NVMbased neuromorphic fabric to mitigate uni-directional retention faults with virtually zero storage overhead depending on the definition of scrub area for multilayer perceptron (MLP) and convolutional neural networks (CNNs). The training of the NNs is adjusted accordingly to meet the requirements of the proposed approximate scrubbing scheme. On different benchmarks, the proposed scrubbing approach can improve the inference accuracy up to 85.51 device operational time with negligible storage overhead.
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:0278-0070
1937-4151
DOI:10.1109/TCAD.2022.3205872