Right for the Right Reasons: Avoiding Reasoning Shortcuts via Prototypical Neurosymbolic AI

TL;DR

This paper introduces Prototypical Neurosymbolic architectures that prevent reasoning shortcuts by ensuring models learn correct concepts, demonstrated through improved performance on synthetic and real-world tasks with limited supervision.

Contribution

The paper proposes a novel prototypical learning approach for neurosymbolic AI to avoid shortcut reasoning, enhancing concept learning in low-data regimes.

Findings

Significant improvements in concept learning on synthetic tasks

Effective avoidance of reasoning shortcuts in real-world scenarios

Enhanced reliability and safety in neurosymbolic models

Abstract

Neurosymbolic AI is growing in popularity thanks to its ability to combine neural perception and symbolic reasoning in end-to-end trainable models. However, recent findings reveal these are prone to shortcut reasoning, i.e., to learning unindented concepts--or neural predicates--which exploit spurious correlations to satisfy the symbolic constraints. In this paper, we address reasoning shortcuts at their root cause and we introduce Prototypical Neurosymbolic architectures. These models are able to satisfy the symbolic constraints (be right) because they have learnt the correct basic concepts (for the right reasons) and not because of spurious correlations, even in extremely low data regimes. Leveraging the theory of prototypical learning, we demonstrate that we can effectively avoid reasoning shortcuts by training the models to satisfy the background knowledge while taking into account the similarity of the input with respect to the handful of labelled datapoints. We extensively validate our approach on the recently proposed rsbench benchmark suite in a variety of settings and tasks with very scarce supervision: we show significant improvements in learning the right concepts both in synthetic tasks (MNIST-EvenOdd and Kand-Logic) and real-world, high-stake ones (BDD-OIA). Our findings pave the way to prototype grounding as an effective, annotation-efficient strategy for safe and reliable neurosymbolic learning.