Reinforcement learning algorithms: A brief survey

• Published in: Expert systems with applications Vol. 231; p. 120495
• Main Authors: Shakya, Ashish Kumar, Pillai, Gopinatha, Chakrabarty, Sohom
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 30.11.2023
• Subjects: Deep Reinforcement Learning (DRL); Function approximation; Reinforcement learning; Stochastic optimal control; Deep Reinforcement Learning (DRL); Function approximation; Stochastic optimal control; Reinforcement learning
• ISSN: 0957-4174; 1873-6793
• DOI: 10.1016/j.eswa.2023.120495

•RL can be used to solve problems involving sequential decision-making.•RL is based on trial-and-error learning through rewards and punishments.•The ultimate goal of an RL agent is to maximize cumulative reward.•RL agent tries to learn the optimal value and policy functions.•DNN-based function approximation is used to approximate the value and policy. Reinforcement Learning (RL) is a machine learning (ML) technique to learn sequential decision-making in complex problems. RL is inspired by trial-and-error based human/animal learning. It can learn an optimal policy autonomously with knowledge obtained by continuous interaction with a stochastic dynamical environment. Problems considered virtually impossible to solve, such as learning to play video games just from pixel information, are now successfully solved using deep reinforcement learning. Without human intervention, RL agents can surpass human performance in challenging tasks. This review gives a broad overview of RL, covering its fundamental principles, essential methods, and illustrative applications. The authors aim to develop an initial reference point for researchers commencing their research work in RL. In this review, the authors cover some fundamental model-free RL algorithms and pathbreaking function approximation-based deep RL (DRL) algorithms for complex uncertain tasks with continuous action and state spaces, making RL useful in various interdisciplinary fields. This article also provides a brief review of model-based and multi-agent RL approaches. Finally, some promising research directions for RL are briefly presented.