Quality of internal representation shapes learning performance in feedback neural networks

TL;DR

This paper investigates how the internal dynamics of feedback neural networks influence their ability to learn sinusoidal signals, revealing that optimal learning occurs at intermediate target frequencies with high-dimensional, noise-robust internal representations.

Contribution

It provides an exact mathematical analysis of linearized networks and demonstrates how internal network parameters and target properties jointly shape learning performance.

Findings

Optimal target frequency decreases with reservoir connectivity strength.

High-dimensional, de-synchronized internal representations are most robust to noise.

Predictions from linear analysis match qualitative behavior of non-linear networks.

Abstract

A fundamental feature of complex biological systems is the ability to form feedback interactions with their environment. A prominent model for studying such interactions is reservoir computing, where learning acts on low-dimensional bottlenecks. Despite the simplicity of this learning scheme, the factors contributing to or hindering the success of training in reservoir networks are in general not well understood. In this work, we study non-linear feedback networks trained to generate a sinusoidal signal, and analyze how learning performance is shaped by the interplay between internal network dynamics and target properties. By performing exact mathematical analysis of linearized networks, we predict that learning performance is maximized when the target is characterized by an optimal, intermediate frequency which monotonically decreases with the strength of the internal reservoir connectivity. At the optimal frequency, the reservoir representation of the target signal is high-dimensional, de-synchronized, and thus maximally robust to noise. We show that our predictions successfully capture the qualitative behaviour of performance in non-linear networks. Moreover, we find that the relationship between internal representations and performance can be further exploited in trained non-linear networks to explain behaviours which do not have a linear counterpart. Our results indicate that a major determinant of learning success is the quality of the internal representation of the target, which in turn is shaped by an interplay between parameters controlling the internal network and those defining the task.