Non-Stationary Learning of Neural Networks with Automatic Soft Parameter Reset

Alexandre Galashov, Michalis Titsias, András György, Clare Lyle, Razvan Pascanu, Yee Whye Teh, Maneesh Sahani

Keywords: Non-stationarity, plasticity loss, online learning, deep learning

TL;DR: Learning to reset Neural Networks parameters using Ornstein-Uhlenbeck to learn on non-stationary data

Abstract: Neural networks are traditionally trained under the assumption that data come from a stationary distribution. However, settings which violate this assumption are becoming more popular; examples include supervised learning under distributional shifts, reinforcement learning, continual learning and non-stationary contextual bandits. In this work we introduce a novel learning approach that automatically models and adapts to non-stationarity, via an Ornstein-Uhlenbeck process with an adaptive drift parameter. The adaptive drift tends to draw the parameters towards the initialisation distribution, so the approach can be understood as a form of soft parameter reset. We show empirically that our approach performs well in non-stationary supervised and off-policy reinforcement learning settings.

Primary Area: Optimization for deep networks

Submission Number: 19536