Declarative Design of Neural Predicates in Neuro-Symbolic Systems

TL;DR

This paper introduces a framework for neural predicates in neuro-symbolic systems that enhances declarative reasoning, enabling flexible query answering without extensive training, thus improving the interpretability and reasoning capabilities of AI systems.

Contribution

It presents a novel, fully declarative neural predicate framework that maintains learning and reasoning abilities in neuro-symbolic systems, addressing a key limitation of existing approaches.

Findings

Framework preserves learning and reasoning capabilities.

Enables answering arbitrary queries after training on a single query type.

Extends neuro-symbolic systems with full declarative properties.

Abstract

Neuro-symbolic systems (NeSy), which claim to combine the best of both learning and reasoning capabilities of artificial intelligence, are missing a core property of reasoning systems: Declarativeness. The lack of declarativeness is caused by the functional nature of neural predicates inherited from neural networks. We propose and implement a general framework for fully declarative neural predicates, which hence extends to fully declarative NeSy frameworks. We first show that the declarative extension preserves the learning and reasoning capabilities while being able to answer arbitrary queries while only being trained on a single query type.