Parameter-Level Soft-Masking for Continual Learning

TL;DR

This paper introduces SPG, a novel parameter-level soft-masking technique for continual learning that prevents catastrophic forgetting, enhances knowledge transfer, and reduces network capacity usage, outperforming existing methods.

Contribution

It is the first work to apply parameter-level soft-masking in continual learning, enabling full network use per task while mitigating forgetting and promoting transfer.

Findings

SPG effectively prevents catastrophic forgetting.

SPG enhances knowledge transfer among similar and dissimilar tasks.

SPG reduces network capacity consumption.

Abstract

Existing research on task incremental learning in continual learning has primarily focused on preventing catastrophic forgetting (CF). Although several techniques have achieved learning with no CF, they attain it by letting each task monopolize a sub-network in a shared network, which seriously limits knowledge transfer (KT) and causes over-consumption of the network capacity, i.e., as more tasks are learned, the performance deteriorates. The goal of this paper is threefold: (1) overcoming CF, (2) encouraging KT, and (3) tackling the capacity problem. A novel technique (called SPG) is proposed that soft-masks (partially blocks) parameter updating in training based on the importance of each parameter to old tasks. Each task still uses the full network, i.e., no monopoly of any part of the network by any task, which enables maximum KT and reduction in capacity usage. To our knowledge, this is the first work that soft-masks a model at the parameter-level for continual learning. Extensive experiments demonstrate the effectiveness of SPG in achieving all three objectives. More notably, it attains significant transfer of knowledge not only among similar tasks (with shared knowledge) but also among dissimilar tasks (with little shared knowledge) while mitigating CF.