Fault-Tolerant Neuromorphic Computing With Memristors Using Functional ATPG for Efficient Recalibration
- 3 May 2023
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Design & Test
- Vol. 40 (4), 42-50
- https://doi.org/10.1109/mdat.2023.3270126
Abstract
Neuromorphic fabric based on emerging resistive non-volatile memories (NVMs) is a promising approach for the realization of Neural Networks (NNs) due to their low power consumption and latency based on in-memory computation. However, the non-idealities of NVMs shift activation distributions from training values, affecting post-mapping inference accuracy. Re-calibrating NNs’ batch normalization layer in a per-chip manner restores a shifted distribution close to the training distribution. However, re-calibration overhead is an issue not addressed in existing studies. Therefore, we propose approximate batch normalization and a test pattern generation method for efficient re-calibration. The proposed method requires only 0.2% of training data for re-calibration, but can regain 72.3% inference accuracy on various benchmark datasets and fault scenarios.Keywords
Funding Information
- German Research Foundation (TA782/29)
This publication has 9 references indexed in Scilit:
- Process and Runtime Variation Robustness for Spintronic-Based Neuromorphic FabricPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2022
- Accurate deep neural network inference using computational phase-change memoryNature Communications, 2020
- Equivalent-accuracy accelerated neural-network training using analogue memoryNature, 2018
- XNOR-RRAM: A scalable and parallel resistive synaptic architecture for binary neural networksPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2018
- Neuro-Inspired Computing With Emerging Nonvolatile MemorysProceedings of the IEEE, 2018
- Rescuing Memristor-based Neuromorphic Design with High DefectsPublished by Association for Computing Machinery (ACM) ,2017
- Causes and consequences of the stochastic aspect of filamentary RRAMMicroelectronic Engineering, 2015
- Redox‐Based Resistive Switching Memories – Nanoionic Mechanisms, Prospects, and ChallengesAdvanced Materials, 2009
- Estimating the mean and variance from the median, range, and the size of a sampleBMC Medical Research Methodology, 2005