Neuro-Symbolic Continual Learning: Knowledge, Reasoning Shortcuts and Concept Rehearsal

TL;DR

This paper presents COOL, a novel continual learning strategy for neuro-symbolic models that effectively preserves high-quality concepts over time, addressing catastrophic forgetting and reasoning shortcuts in sequential tasks.

Contribution

The paper introduces COOL, a concept-level continual learning method that improves knowledge retention and reasoning accuracy in neuro-symbolic systems.

Findings

COOL outperforms existing strategies on three new benchmarks.

It effectively prevents reasoning shortcuts and maintains concept semantics.

The approach achieves sustained high performance in neuro-symbolic continual learning tasks.

Abstract

We introduce Neuro-Symbolic Continual Learning, where a model has to solve a sequence of neuro-symbolic tasks, that is, it has to map sub-symbolic inputs to high-level concepts and compute predictions by reasoning consistently with prior knowledge. Our key observation is that neuro-symbolic tasks, although different, often share concepts whose semantics remains stable over time. Traditional approaches fall short: existing continual strategies ignore knowledge altogether, while stock neuro-symbolic architectures suffer from catastrophic forgetting. We show that leveraging prior knowledge by combining neuro-symbolic architectures with continual strategies does help avoid catastrophic forgetting, but also that doing so can yield models affected by reasoning shortcuts. These undermine the semantics of the acquired concepts, even when detailed prior knowledge is provided upfront and inference is exact, and in turn continual performance. To overcome these issues, we introduce COOL, a COncept-level cOntinual Learning strategy tailored for neuro-symbolic continual problems that acquires high-quality concepts and remembers them over time. Our experiments on three novel benchmarks highlights how COOL attains sustained high performance on neuro-symbolic continual learning tasks in which other strategies fail.