Introduction In the era of big data, the fields of artificial intelligence and machine learning provide powerful tools and techniques in a broad range of applications for solving large-scale data-driven practical problems. Reinforcement learning (RL) is one of the major driving forces that enable machine learning to learn skills beyond the human level through interacting with complex environments. In recent years, with the development of new theories and algorithms, RL has achieved tremendous successes in many commercialized applications, including autonomous driving, recommendation systems, wireless communications, robotics, and gaming. Nevertheless, RL is still a largely underexplored area with many open problems waiting to be addressed, including fundamental RL theory, RL algorithm design, and application of RL to other multi-disciplinary areas.

This workshop aims to cover a broad spectrum of RL topics, including, but not limited to, foundational theory, algorithm design, multi-disciplinary applications, robustness, security, privacy and applications. This workshop also aims to bring together researchers from multi-disciplinary areas and help identify potential directions and applications for future developments of RL.

Important Dates

• Oct 15, 2021: Due date for full workshop papers submission

• Nov 1, 2021: Notification of paper acceptance to authors

• Nov 20, 2021: Camera-ready of accepted papers

• Dec 15-18, 2021: Workshops

Topics of Interest (Non-comprehensive List)

• RL theory and algorithms

• Multi-agent, distributed RL

• Transfer and meta RL

• Federated RL

• Robust, cyber-secure RL

• RL with (physical) constraints

• Privacy in RL

• Offline RL

• Model-based and model-free RL

• Optimization approaches in RL

• Emerging RL applications

• Inverse RL and Imitation Learning

• RL in industry, autonomous driving

• Causal RL