Meta-Learning Representations for Continual Learning

TL;DR

This paper introduces OML, a novel objective for continual learning that learns sparse, robust representations to reduce forgetting and accelerate future learning, outperforming some rehearsal-based methods.

Contribution

It proposes a new learning objective, OML, that directly minimizes catastrophic interference and enhances online learning in continual learning settings.

Findings

OML learns sparse, effective representations for online updates.

OML is compatible with existing continual learning strategies like MER and GEM.

A simple online update with OML rivals rehearsal-based methods in performance.

Abstract

A continual learning agent should be able to build on top of existing knowledge to learn on new data quickly while minimizing forgetting. Current intelligent systems based on neural network function approximators arguably do the opposite---they are highly prone to forgetting and rarely trained to facilitate future learning. One reason for this poor behavior is that they learn from a representation that is not explicitly trained for these two goals. In this paper, we propose OML, an objective that directly minimizes catastrophic interference by learning representations that accelerate future learning and are robust to forgetting under online updates in continual learning. We show that it is possible to learn naturally sparse representations that are more effective for online updating. Moreover, our algorithm is complementary to existing continual learning strategies, such as MER and GEM. Finally, we demonstrate that a basic online updating strategy on representations learned by OML is competitive with rehearsal based methods for continual learning. We release an implementation of our method at https://github.com/khurramjaved96/mrcl .