puc-header

Continual Learning in a Multi-Layer Network of an Electric Fish

30 Pages Posted: 24 May 2019 Sneak Peek Status: Review Complete

See all articles by Salomon Z. Muller

Salomon Z. Muller

Columbia University - Zuckerman Mind Brain Behavior Institute

Abigail Zadina

Columbia University - Zuckerman Mind Brain Behavior Institute

L.F. Abbott

Columbia University - Zuckerman Mind Brain Behavior Institute

Nate Sawtell

Columbia University - Zuckerman Mind Brain Behavior Institute

More...

Abstract

Distributing learning across multiple layers has proven extremely powerful in artificial neural networks. However, little is known about how such multi-layer learning is implemented in the brain. Here we provide a detailed account of multi-layer learning in the electrosensory lobe (ELL) of mormyrid fish, a tractable biological model system, and report how two longstanding problems in multi-layer learning are solved. First, we find that the problem of separating learning signals from output signals suitable for driving behavior is solved by a functional compartmentalization, in which inputs driving learning differentially affect dendritic and axonal spikes in intermediate layer neurons. Second, we find that error 'credit assignment' is solved by organizing the connectivity between intermediate and output layer neurons on the basis of learning rather than sensory response selectivity. These mechanisms have broad relevance to learning in the cerebellum, hippocampus and cerebral cortex as well as in artificial systems.

Keywords: neural circuits, learning, synaptic plasticity, machine learning, corollary discharge, electric fish

Suggested Citation

Muller, Salomon Z. and Zadina, Abigail and Abbott, L.F. and Sawtell, Nate, Continual Learning in a Multi-Layer Network of an Electric Fish (May 24, 2019). CELL-D-19-01360. Available at SSRN: https://ssrn.com/abstract=3393520 or http://dx.doi.org/10.2139/ssrn.3393520
This is a paper under consideration at Cell Press and has not been peer-reviewed.

Salomon Z. Muller

Columbia University - Zuckerman Mind Brain Behavior Institute

United States

Abigail Zadina

Columbia University - Zuckerman Mind Brain Behavior Institute

United States

L.F. Abbott

Columbia University - Zuckerman Mind Brain Behavior Institute

United States

Nate Sawtell (Contact Author)

Columbia University - Zuckerman Mind Brain Behavior Institute ( email )

United States

Click here to go to Cell.com

Go to Cell.com

Paper statistics

Abstract Views
1,045
Downloads
46