R5: Rule Discovery with Reinforced and Recurrent Relational Reasoning
This addresses the challenge of enabling machines to generalize from small-scale to large-scale tasks through explainable rule discovery, which is incremental as it builds on existing relational reasoning methods.
The paper tackles the problem of systematicity in relational reasoning by proposing R5, a reinforcement learning framework that mines compositional logical rules from relational graph data, achieving superior performance on relation prediction tasks and high recall in discovering ground truth rules.
Systematicity, i.e., the ability to recombine known parts and rules to form new sequences while reasoning over relational data, is critical to machine intelligence. A model with strong systematicity is able to train on small-scale tasks and generalize to large-scale tasks. In this paper, we propose R5, a relational reasoning framework based on reinforcement learning that reasons over relational graph data and explicitly mines underlying compositional logical rules from observations. R5 has strong systematicity and being robust to noisy data. It consists of a policy value network equipped with Monte Carlo Tree Search to perform recurrent relational prediction and a backtrack rewriting mechanism for rule mining. By alternately applying the two components, R5 progressively learns a set of explicit rules from data and performs explainable and generalizable relation prediction. We conduct extensive evaluations on multiple datasets. Experimental results show that R5 outperforms various embedding-based and rule induction baselines on relation prediction tasks while achieving a high recall rate in discovering ground truth rules.