Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

RMP-Loss: Regularizing Membrane Potential Distribution for Spiking Neural Networks

About

Spiking Neural Networks (SNNs) as one of the biology-inspired models have received much attention recently. It can significantly reduce energy consumption since they quantize the real-valued membrane potentials to 0/1 spikes to transmit information thus the multiplications of activations and weights can be replaced by additions when implemented on hardware. However, this quantization mechanism will inevitably introduce quantization error, thus causing catastrophic information loss. To address the quantization error problem, we propose a regularizing membrane potential loss (RMP-Loss) to adjust the distribution which is directly related to quantization error to a range close to the spikes. Our method is extremely simple to implement and straightforward to train an SNN. Furthermore, it is shown to consistently outperform previous state-of-the-art methods over different network architectures and datasets.

Yufei Guo, Xiaode Liu, Yuanpei Chen, Liwen Zhang, Weihang Peng, Yuhan Zhang, Xuhui Huang, Zhe Ma• 2023

Related benchmarks

TaskDatasetResultRank
Image ClassificationCIFAR-10--
973
Image ClassificationImageNet 1k (test)
Top-1 Accuracy65.17
939
Image ClassificationCIFAR-100--
375
Image ClassificationCIFAR-100
Accuracy78.28
302
Image ClassificationCIFAR-10
Accuracy93.33
23
Image ClassificationCIFAR-100
Accuracy72.55
23
Image ClassificationImageNet
Accuracy (T=250)63.03
18
Showing 7 of 7 rows

Other info

Follow for update