GAN Q-learning

2018/05/13 by Thang Doan, Doan, Thang, Bogdan Mazoure +3
Computer Science · #Adversarial Robustness in Machine Learning #Anomaly Detection Techniques and Applications #FOS: Computer and information sciences #Machine Learning (cs.LG) #Machine Learning (stat.ML) #Reinforcement Learning in Robotics

paper · pdf · doi:10.48550/arxiv.1805.04874

openalex publication_date 2018/05/13 · openalex created_date 2022/10/01 · openalex updated_date 2026/07/28

Abstract

Distributional reinforcement learning (distributional RL) has seen empirical success in complex Markov Decision Processes (MDPs) in the setting of nonlinear function approximation. However, there are many different ways in which one can leverage the distributional approach to reinforcement learning. In this paper, we propose GAN Q-learning, a novel distributional RL method based on generative adversarial networks (GANs) and analyze its performance in simple tabular environments, as well as OpenAI Gym. We empirically show that our algorithm leverages the flexibility and blackbox approach of deep learning models while providing a viable alternative to traditional methods.

Related

• Non-decreasing Quantile Function Network with Efficient Exploration for Distributional Reinforcement Learning
• Distributional Multivariate Policy Evaluation and Exploration with the\n Bellman GAN
• Knowledge Transfer in Deep Reinforcement Learning via an RL-Specific GAN-Based Correspondence Function
• Efficient Exploration through Bayesian Deep Q-Networks
• Analytically Tractable Bayesian Deep Q-Learning