Neuro-symbolic Rule Learning in Real-world Classification Tasks

TL;DR

This paper extends neuro-symbolic rule learning with neural DNF models to handle real-world multi-class and multi-label classification tasks, improving interpretability and scalability over prior synthetic binary classification applications.

Contribution

It introduces an extended neural DNF model called neural DNF-EO that enforces mutual exclusivity in multi-class classification and demonstrates improved scalability and interpretability.

Findings

Neural DNF models perform comparably to neural networks in accuracy.

Neural DNF models provide better interpretability through logical rule extraction.

Neural DNF models scale better than FastLAS in large multi-class and multi-label tasks.

Abstract

Neuro-symbolic rule learning has attracted lots of attention as it offers better interpretability than pure neural models and scales better than symbolic rule learning. A recent approach named pix2rule proposes a neural Disjunctive Normal Form (neural DNF) module to learn symbolic rules with feed-forward layers. Although proved to be effective in synthetic binary classification, pix2rule has not been applied to more challenging tasks such as multi-label and multi-class classifications over real-world data. In this paper, we address this limitation by extending the neural DNF module to (i) support rule learning in real-world multi-class and multi-label classification tasks, (ii) enforce the symbolic property of mutual exclusivity (i.e. predicting exactly one class) in multi-class classification, and (iii) explore its scalability over large inputs and outputs. We train a vanilla neural DNF model similar to pix2rule's neural DNF module for multi-label classification, and we propose a novel extended model called neural DNF-EO (Exactly One) which enforces mutual exclusivity in multi-class classification. We evaluate the classification performance, scalability and interpretability of our neural DNF-based models, and compare them against pure neural models and a state-of-the-art symbolic rule learner named FastLAS. We demonstrate that our neural DNF-based models perform similarly to neural networks, but provide better interpretability by enabling the extraction of logical rules. Our models also scale well when the rule search space grows in size, in contrast to FastLAS, which fails to learn in multi-class classification tasks with 200 classes and in all multi-label settings.