Theories of synaptic memory consolidation and intelligent plasticity for continual learning

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Zenke, Friedemann, Laborieux, Axel
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866929548796362752
author Zenke, Friedemann
Laborieux, Axel
author_facet Zenke, Friedemann
Laborieux, Axel
contents Humans and animals learn throughout life. Such continual learning is crucial for intelligence. In this chapter, we examine the pivotal role plasticity mechanisms with complex internal synaptic dynamics could play in enabling this ability in neural networks. By surveying theoretical research, we highlight two fundamental enablers for continual learning. First, synaptic plasticity mechanisms must maintain and evolve an internal state over several behaviorally relevant timescales. Second, plasticity algorithms must leverage the internal state to intelligently regulate plasticity at individual synapses to facilitate the seamless integration of new memories while avoiding detrimental interference with existing ones. Our chapter covers successful applications of these principles to deep neural networks and underscores the significance of synaptic metaplasticity in sustaining continual learning capabilities. Finally, we outline avenues for further research to understand the brain's superb continual learning abilities and harness similar mechanisms for artificial intelligence systems.
format Preprint
id arxiv_https___arxiv_org_abs_2405_16922
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Theories of synaptic memory consolidation and intelligent plasticity for continual learning
Zenke, Friedemann
Laborieux, Axel
Neurons and Cognition
Artificial Intelligence
Machine Learning
Neural and Evolutionary Computing
Humans and animals learn throughout life. Such continual learning is crucial for intelligence. In this chapter, we examine the pivotal role plasticity mechanisms with complex internal synaptic dynamics could play in enabling this ability in neural networks. By surveying theoretical research, we highlight two fundamental enablers for continual learning. First, synaptic plasticity mechanisms must maintain and evolve an internal state over several behaviorally relevant timescales. Second, plasticity algorithms must leverage the internal state to intelligently regulate plasticity at individual synapses to facilitate the seamless integration of new memories while avoiding detrimental interference with existing ones. Our chapter covers successful applications of these principles to deep neural networks and underscores the significance of synaptic metaplasticity in sustaining continual learning capabilities. Finally, we outline avenues for further research to understand the brain's superb continual learning abilities and harness similar mechanisms for artificial intelligence systems.
title Theories of synaptic memory consolidation and intelligent plasticity for continual learning
topic Neurons and Cognition
Artificial Intelligence
Machine Learning
Neural and Evolutionary Computing
url https://arxiv.org/abs/2405.16922