Neurosymbolic Reinforcement Learning and Planning: A Survey

TL;DR

This survey reviews the rapidly evolving field of Neurosymbolic Reinforcement Learning, categorizing existing works, analyzing their components, and highlighting future research opportunities and challenges.

Contribution

It provides a comprehensive taxonomy and analysis of Neurosymbolic RL research, focusing on neural, symbolic, and RL components and their roles.

Findings

Categorized works into three main types: Learning for Reasoning, Reasoning for Learning, and Learning-Reasoning.

Analyzed RL components such as state space, action space, policy, and algorithms across studies.

Identified key research challenges and future directions in Neurosymbolic RL applications.

Abstract

The area of Neurosymbolic Artificial Intelligence (Neurosymbolic AI) is rapidly developing and has become a popular research topic, encompassing sub-fields such as Neurosymbolic Deep Learning (Neurosymbolic DL) and Neurosymbolic Reinforcement Learning (Neurosymbolic RL). Compared to traditional learning methods, Neurosymbolic AI offers significant advantages by simplifying complexity and providing transparency and explainability. Reinforcement Learning(RL), a long-standing Artificial Intelligence(AI) concept that mimics human behavior using rewards and punishment, is a fundamental component of Neurosymbolic RL, a recent integration of the two fields that has yielded promising results. The aim of this paper is to contribute to the emerging field of Neurosymbolic RL by conducting a literature survey. Our evaluation focuses on the three components that constitute Neurosymbolic RL: neural, symbolic, and RL. We categorize works based on the role played by the neural and symbolic parts in RL, into three taxonomies:Learning for Reasoning, Reasoning for Learning and Learning-Reasoning. These categories are further divided into sub-categories based on their applications. Furthermore, we analyze the RL components of each research work, including the state space, action space, policy module, and RL algorithm. Additionally, we identify research opportunities and challenges in various applications within this dynamic field.