Neural-Symbolic Commonsense Reasoner with Relation Predictors

TL;DR

This paper introduces a neural-symbolic reasoning model for commonsense knowledge graphs that learns logic rules during training, enabling better reasoning over unseen events and outperforming existing models.

Contribution

The proposed model learns logic rules for reasoning over large-scale, dynamic CKGs, improving generalization and interpretability compared to prior approaches.

Findings

Outperforms state-of-the-art models on link prediction tasks

Learns interpretable logic rules during training

Effectively generalizes to unseen events

Abstract

Commonsense reasoning aims to incorporate sets of commonsense facts, retrieved from Commonsense Knowledge Graphs (CKG), to draw conclusion about ordinary situations. The dynamic nature of commonsense knowledge postulates models capable of performing multi-hop reasoning over new situations. This feature also results in having large-scale sparse Knowledge Graphs, where such reasoning process is needed to predict relations between new events. However, existing approaches in this area are limited by considering CKGs as a limited set of facts, thus rendering them unfit for reasoning over new unseen situations and events. In this paper, we present a neural-symbolic reasoner, which is capable of reasoning over large-scale dynamic CKGs. The logic rules for reasoning over CKGs are learned during training by our model. In addition to providing interpretable explanation, the learned logic rules help to generalise prediction to newly introduced events. Experimental results on the task of link prediction on CKGs prove the effectiveness of our model by outperforming the state-of-the-art models.