Neural Decision-Propagation for Answer Set Programming

2026-05-03

AI Summary

This work addresses the scalability limitations of traditional neuro-symbolic systems that rely on classical solvers to compute stable models in Answer Set Programming (ASP). The authors propose Decision Propagation (DProp), an algorithm that efficiently computes stable models by alternately performing false-value determination and true-value propagation. They further introduce NDProp, a differentiable extension of DProp that, for the first time, integrates differentiable neural computation into ASP solving. NDProp combines neural decision-making with fuzzy logical propagation to enable end-to-end neuro-symbolic reasoning without requiring classical solvers. Experimental results demonstrate that the approach effectively learns decision heuristics and significantly improves both accuracy and scalability on standard neuro-symbolic benchmarks.

Abstract

Integration of Answer Set Programming (ASP) with neural networks has emerged as a promising tool in Neuro-symbolic AI. While existing approaches extend the capabilities of ASP to real world domains, their reasoning pipelines depend on classical solvers, which is a bottleneck for scalability. To tackle this problem, we propose a new method to compute stable models, called decision-propagation (DProp), which alternates falsity decisions and truth propagations. Successful DProp computations are shown to capture the stable model semantics. We then develop Neural DProp (NDProp), a differentiable extension of DProp with neural computation for decisions and fuzzy evaluation for propagations. We evaluate the capabilities of NDProp for learning decision heuristics as well as neuro-symbolic integration, and compare it with existing neuro-symbolic approaches. The results show that NDProp can learn to efficiently compute stable models, and it improves accuracy and scalability on neuro-symbolic benchmarks.

Problem

• Answer Set Programming
• Neuro-symbolic AI
• Scalability
• Stable Models
• Reasoning

Innovation

• Neural Answer Set Programming
• Decision Propagation
• Neuro-symbolic AI
• Differentiable Reasoning
• Stable Model Semantics