Defeating Catastrophic Forgetting via Enhanced Orthogonal Weights Modification

TL;DR

This paper introduces EOWM, an enhanced orthogonal weights modification method, to effectively address catastrophic forgetting in neural networks by analyzing weight gradients, proposing a new strategy, and establishing an upper bound on learnable tasks, with superior experimental results.

Contribution

The paper proposes EOWM, a novel continual learning method that enhances OWM by analyzing weight gradients and establishing an upper bound on learnable tasks.

Findings

EOWM outperforms state-of-the-art CL methods on benchmarks.

The weight gradient depends on input space and previous weights.

An upper bound for sequential task learning is theoretically established.

Abstract

The ability of neural networks (NNs) to learn and remember multiple tasks sequentially is facing tough challenges in achieving general artificial intelligence due to their catastrophic forgetting (CF) issues. Fortunately, the latest OWM Orthogonal Weights Modification) and other several continual learning (CL) methods suggest some promising ways to overcome the CF issue. However, none of existing CL methods explores the following three crucial questions for effectively overcoming the CF issue: that is, what knowledge does it contribute to the effective weights modification of the NN during its sequential tasks learning? When the data distribution of a new learning task changes corresponding to the previous learned tasks, should a uniform/specific weight modification strategy be adopted or not? what is the upper bound of the learningable tasks sequentially for a given CL method? ect. To achieve this, in this paper, we first reveals the fact that of the weight gradient of a new learning task is determined by both the input space of the new task and the weight space of the previous learned tasks sequentially. On this observation and the recursive least square optimal method, we propose a new efficient and effective continual learning method EOWM via enhanced OWM. And we have theoretically and definitively given the upper bound of the learningable tasks sequentially of our EOWM. Extensive experiments conducted on the benchmarks demonstrate that our EOWM is effectiveness and outperform all of the state-of-the-art CL baselines.