Progressive Learning without Forgetting

TL;DR

This paper introduces Progressive Learning without Forgetting (PLwF), a continual learning method that prevents catastrophic forgetting and balances stability and plasticity without using old data.

Contribution

The paper proposes PLwF and a credit assignment regime that effectively mitigate forgetting and manage learning dynamics in continual learning without relying on previous data.

Findings

PLwF outperforms existing methods in continual learning tasks.

The credit assignment regime reduces gradient conflicts.

The approach works without access to old data.

Abstract

Learning from changing tasks and sequential experience without forgetting the obtained knowledge is a challenging problem for artificial neural networks. In this work, we focus on two challenging problems in the paradigm of Continual Learning (CL) without involving any old data: (i) the accumulation of catastrophic forgetting caused by the gradually fading knowledge space from which the model learns the previous knowledge; (ii) the uncontrolled tug-of-war dynamics to balance the stability and plasticity during the learning of new tasks. In order to tackle these problems, we present Progressive Learning without Forgetting (PLwF) and a credit assignment regime in the optimizer. PLwF densely introduces model functions from previous tasks to construct a knowledge space such that it contains the most reliable knowledge on each task and the distribution information of different tasks, while credit assignment controls the tug-of-war dynamics by removing gradient conflict through projection. Extensive ablative experiments demonstrate the effectiveness of PLwF and credit assignment. In comparison with other CL methods, we report notably better results even without relying on any raw data.