Gated Linear Networks
Joel Veness, Tor Lattimore, David Budden, Avishkar Bhoopchand, Christopher Mattern, Agnieszka Grabska-Barwinska, Eren Sezener, Jianan Wang
Abstract
This paper presents a new family of backpropagation-free neural architectures, Gated Linear Networks (GLNs). What distinguishes GLNs from contemporary neural networks is the distributed and local nature of their credit assignment mechanism; each neuron directly predicts the target, forgoing the ability to learn feature representations in favor of rapid online learning. Individual neurons are able to model nonlinear functions via the use of data-dependent gating in conjunction with online convex optimization. We show that this architecture gives rise to universal learning capabilities in the limit, with effective model capacity increasing as a function of network size in a manner comparable with deep ReLU networks. Furthermore, we demonstrate that the GLN learning mechanism possesses extraordinary resilience to catastrophic forgetting, performing almost on par to an MLP with dropout and Elastic Weight Consolidation on standard benchmarks.
BibTeX
@inproceedings{aaai2021_gatedlinearnetwo,
title = {Gated Linear Networks},
author = {Joel Veness and Tor Lattimore and David Budden and Avishkar Bhoopchand and Christopher Mattern and Agnieszka Grabska-Barwinska and Eren Sezener and Jianan Wang and Peter Toth and Simon Schmitt and Marcus Hutter},
booktitle = {AAAI 2021},
year = {2021}
}