Metalearning Continual Learning Algorithms

TL;DR

This paper introduces Automated Continual Learning (ACL), a meta-learning approach enabling neural networks to self-develop algorithms that effectively prevent catastrophic forgetting in continual learning scenarios.

Contribution

It proposes a novel meta-learning framework that trains neural networks to generate in-context continual learning algorithms, addressing catastrophic forgetting without hand-crafted solutions.

Findings

ACL outperforms hand-crafted algorithms on Split-MNIST

ACL effectively mitigates in-context catastrophic forgetting

Enables continual learning across multiple image datasets

Abstract

General-purpose learning systems should improve themselves in open-ended fashion in ever-changing environments. Conventional learning algorithms for neural networks, however, suffer from catastrophic forgetting (CF), i.e., previously acquired skills are forgotten when a new task is learned. Instead of hand-crafting new algorithms for avoiding CF, we propose Automated Continual Learning (ACL) to train self-referential neural networks to metalearn their own in-context continual (meta)learning algorithms. ACL encodes continual learning (CL) desiderata -- good performance on both old and new tasks -- into its metalearning objectives. Our experiments demonstrate that ACL effectively resolves "in-context catastrophic forgetting," a problem that naive in-context learning algorithms suffer from; ACL-learned algorithms outperform both hand-crafted learning algorithms and popular meta-continual learning methods on the Split-MNIST benchmark in the replay-free setting, and enables continual learning of diverse tasks consisting of multiple standard image classification datasets. We also discuss the current limitations of in-context CL by comparing ACL with state-of-the-art CL methods that leverage pre-trained models. Overall, we bring several novel perspectives into the long-standing problem of CL.