Spike Accumulation Forwarding for Effective Training of Spiking Neural Networks

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Saiin, Ryuji, Shirakawa, Tomoya, Yoshihara, Sota, Sawada, Yoshihide, Kusumoto, Hiroyuki
Format: Preprint
Published: 2023
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866929402425638912
author Saiin, Ryuji
Shirakawa, Tomoya
Yoshihara, Sota
Sawada, Yoshihide
Kusumoto, Hiroyuki
author_facet Saiin, Ryuji
Shirakawa, Tomoya
Yoshihara, Sota
Sawada, Yoshihide
Kusumoto, Hiroyuki
contents In this article, we propose a new paradigm for training spiking neural networks (SNNs), spike accumulation forwarding (SAF). It is known that SNNs are energy-efficient but difficult to train. Consequently, many researchers have proposed various methods to solve this problem, among which online training through time (OTTT) is a method that allows inferring at each time step while suppressing the memory cost. However, to compute efficiently on GPUs, OTTT requires operations with spike trains and weighted summation of spike trains during forwarding. In addition, OTTT has shown a relationship with the Spike Representation, an alternative training method, though theoretical agreement with Spike Representation has yet to be proven. Our proposed method can solve these problems; namely, SAF can halve the number of operations during the forward process, and it can be theoretically proven that SAF is consistent with the Spike Representation and OTTT, respectively. Furthermore, we confirmed the above contents through experiments and showed that it is possible to reduce memory and training time while maintaining accuracy.
format Preprint
id arxiv_https___arxiv_org_abs_2310_02772
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Spike Accumulation Forwarding for Effective Training of Spiking Neural Networks
Saiin, Ryuji
Shirakawa, Tomoya
Yoshihara, Sota
Sawada, Yoshihide
Kusumoto, Hiroyuki
Neural and Evolutionary Computing
Artificial Intelligence
In this article, we propose a new paradigm for training spiking neural networks (SNNs), spike accumulation forwarding (SAF). It is known that SNNs are energy-efficient but difficult to train. Consequently, many researchers have proposed various methods to solve this problem, among which online training through time (OTTT) is a method that allows inferring at each time step while suppressing the memory cost. However, to compute efficiently on GPUs, OTTT requires operations with spike trains and weighted summation of spike trains during forwarding. In addition, OTTT has shown a relationship with the Spike Representation, an alternative training method, though theoretical agreement with Spike Representation has yet to be proven. Our proposed method can solve these problems; namely, SAF can halve the number of operations during the forward process, and it can be theoretically proven that SAF is consistent with the Spike Representation and OTTT, respectively. Furthermore, we confirmed the above contents through experiments and showed that it is possible to reduce memory and training time while maintaining accuracy.
title Spike Accumulation Forwarding for Effective Training of Spiking Neural Networks
topic Neural and Evolutionary Computing
Artificial Intelligence
url https://arxiv.org/abs/2310.02772