Interpretable Model-based Hierarchical Reinforcement Learning using Inductive Logic Programming

TL;DR

This paper introduces a hierarchical reinforcement learning framework that uses inductive logic programming to improve data efficiency and interpretability of policies, making RL models more transparent and understandable.

Contribution

The novel framework combines symbolic RL with ILP to enhance data efficiency and interpretability without prior knowledge of state transitions.

Findings

Achieved 30-40% improvement in data efficiency over previous methods.

Utilized ILP to learn interpretable symbolic transition rules.

Framework enhances transparency and understanding of learned policies.

Abstract

Recently deep reinforcement learning has achieved tremendous success in wide ranges of applications. However, it notoriously lacks data-efficiency and interpretability. Data-efficiency is important as interacting with the environment is expensive. Further, interpretability can increase the transparency of the black-box-style deep RL models and hence gain trust from the users. In this work, we propose a new hierarchical framework via symbolic RL, leveraging a symbolic transition model to improve the data-efficiency and introduce the interpretability for learned policy. This framework consists of a high-level agent, a subtask solver and a symbolic transition model. Without assuming any prior knowledge on the state transition, we adopt inductive logic programming (ILP) to learn the rules of symbolic state transitions, introducing interpretability and making the learned behavior understandable to users. In empirical experiments, we confirmed that the proposed framework offers approximately between 30\% to 40\% more data efficiency over previous methods.