Smooth Exact Gradient Descent Learning in Spiking Neural Networks

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Klos, Christian, Memmesheimer, Raoul-Martin
Format: Preprint
Published: 2023
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912205902970880
author Klos, Christian
Memmesheimer, Raoul-Martin
author_facet Klos, Christian
Memmesheimer, Raoul-Martin
contents Gradient descent prevails in artificial neural network training, but seems inept for spiking neural networks as small parameter changes can cause sudden, disruptive (dis-)appearances of spikes. Here, we demonstrate exact gradient descent based on continuously changing spiking dynamics. These are generated by neuron models whose spikes vanish and appear at the end of a trial, where it cannot influence subsequent dynamics. This also enables gradient-based spike addition and removal. We illustrate our scheme with various tasks and setups, including recurrent and deep, initially silent networks.
format Preprint
id arxiv_https___arxiv_org_abs_2309_14523
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Smooth Exact Gradient Descent Learning in Spiking Neural Networks
Klos, Christian
Memmesheimer, Raoul-Martin
Neurons and Cognition
Neural and Evolutionary Computing
Gradient descent prevails in artificial neural network training, but seems inept for spiking neural networks as small parameter changes can cause sudden, disruptive (dis-)appearances of spikes. Here, we demonstrate exact gradient descent based on continuously changing spiking dynamics. These are generated by neuron models whose spikes vanish and appear at the end of a trial, where it cannot influence subsequent dynamics. This also enables gradient-based spike addition and removal. We illustrate our scheme with various tasks and setups, including recurrent and deep, initially silent networks.
title Smooth Exact Gradient Descent Learning in Spiking Neural Networks
topic Neurons and Cognition
Neural and Evolutionary Computing
url https://arxiv.org/abs/2309.14523