Discrete linear-complexity reinforcement learning in continuous action\n spaces for Q-learning algorithms

2018/07/16 by Peyman Tavallali, Tavallali, Peyman, Gary Doran +3
Computer Science · Engineering · #Advanced Memory and Neural Computing #Artificial Intelligence (cs.AI) #Evolutionary Algorithms and Applications #FOS: Computer and information sciences #Machine Learning (cs.LG) #Machine Learning (stat.ML) #Neural Networks and Applications #Reinforcement Learning in Robotics

paper · pdf · doi:10.48550/arxiv.1807.06957

openalex publication_date 2018/07/16 · openalex created_date 2022/10/01 · openalex updated_date 2026/07/28

Abstract

In this article, we sketch an algorithm that extends the Q-learning\nalgorithms to the continuous action space domain. Our method is based on the\ndiscretization of the action space. Despite the commonly used discretization\nmethods, our method does not increase the discretized problem dimensionality\nexponentially. We will show that our proposed method is linear in complexity\nwhen the discretization is employed. The variant of the Q-learning algorithm\npresented in this work, labeled as Finite Step Q-Learning (FSQ), can be\ndeployed to both shallow and deep neural network architectures.\n

Citations

• Continuous control with deep reinforcement learning
• Prioritized Experience Replay

Related

• Parametrized Deep Q-Networks Learning: Reinforcement Learning with Discrete-Continuous Hybrid Action Space
• Deep Reinforcement Learning with Adjustments
• Characterizing the Action-Generalization Gap in Deep Q-Learning
• Q-Networks for Binary Vector Actions
• Deep Reinforcement Learning in Large Discrete Action Spaces