Weakly Supervised Reasoning by Neuro-Symbolic Approaches

TL;DR

This paper discusses neuro-symbolic methods that combine AI paradigms to improve interpretability and reasoning in NLP tasks through weak supervision, integrating symbolic structures with neural networks.

Contribution

It introduces a framework that integrates symbolic latent structures with neural models for weakly supervised reasoning in NLP, demonstrating versatility across multiple tasks.

Findings

Successful application to table query reasoning

Effective in syntactic structure reasoning

Improved performance in information extraction

Abstract

Deep learning has largely improved the performance of various natural language processing (NLP) tasks. However, most deep learning models are black-box machinery, and lack explicit interpretation. In this chapter, we will introduce our recent progress on neuro-symbolic approaches to NLP, which combines different schools of AI, namely, symbolism and connectionism. Generally, we will design a neural system with symbolic latent structures for an NLP task, and apply reinforcement learning or its relaxation to perform weakly supervised reasoning in the downstream task. Our framework has been successfully applied to various tasks, including table query reasoning, syntactic structure reasoning, information extraction reasoning, and rule reasoning. For each application, we will introduce the background, our approach, and experimental results.