AnalogNAS-Bench extends NAS-Bench-201 with analog in-memory computing metrics, revealing that architecture rankings under quantization do not transfer to analog noise, and that 3x3 convolutions, pooling, and skip connections aid robustness.
Accurate deep neural network inference using computational phase-change memory
1 Pith paper cite this work. Polarity classification is still indexing.
abstract
In-memory computing is a promising non-von Neumann approach for making energy-efficient deep learning inference hardware. Crossbar arrays of resistive memory devices can be used to encode the network weights and perform efficient analog matrix-vector multiplications without intermediate movements of data. However, due to device variability and noise, the network needs to be trained in a specific way so that transferring the digitally trained weights to the analog resistive memory devices will not result in significant loss of accuracy. Here, we introduce a methodology to train ResNet-type convolutional neural networks that results in no appreciable accuracy loss when transferring weights to in-memory computing hardware based on phase-change memory (PCM). We also propose a compensation technique that exploits the batch normalization parameters to improve the accuracy retention over time. We achieve a classification accuracy of 93.7% on the CIFAR-10 dataset and a top-1 accuracy on the ImageNet benchmark of 71.6% after mapping the trained weights to PCM. Our hardware results on CIFAR-10 with ResNet-32 demonstrate an accuracy above 93.5% retained over a one day period, where each of the 361,722 synaptic weights of the network is programmed on just two PCM devices organized in a differential configuration.
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
AnalogNAS-Bench: A NAS Benchmark for Analog In-Memory Computing
AnalogNAS-Bench extends NAS-Bench-201 with analog in-memory computing metrics, revealing that architecture rankings under quantization do not transfer to analog noise, and that 3x3 convolutions, pooling, and skip connections aid robustness.