Using Fast Weights to Attend to the Recent Past

Until recently, research on artificial neural networks was largely restricted to systems with only two types of variable: Neural activities that represent the current or recent input and weights that learn to capture regularities among inputs, outputs and payoffs. There is no good reason for this re...

Full description

Saved in:
Bibliographic Details
Published in:arXiv.org 2016-12
Main Authors: Ba, Jimmy, Hinton, Geoffrey, Mnih, Volodymyr, Leibo, Joel Z, Ionescu, Catalin
Format: Article
Language:English
Subjects:
Artificial neural networks
Neural networks
Synapses
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Until recently, research on artificial neural networks was largely restricted to systems with only two types of variable: Neural activities that represent the current or recent input and weights that learn to capture regularities among inputs, outputs and payoffs. There is no good reason for this restriction. Synapses have dynamics at many different time-scales and this suggests that artificial neural networks might benefit from variables that change slower than activities but much faster than the standard weights. These "fast weights" can be used to store temporary memories of the recent past and they provide a neurally plausible way of implementing the type of attention to the past that has recently proved very helpful in sequence-to-sequence models. By using fast weights we can avoid the need to store copies of neural activity patterns.
ISSN:2331-8422