The field of reinforcement learning and multi-agent systems is rapidly moving towards more complex and realistic applications. One of the key directions is the development of more efficient and scalable algorithms for large-scale problems. Recent papers have introduced novel frameworks for parallel reinforcement learning, such as maximizing state entropy in parallel settings, and distributed hierarchical reinforcement learning approaches based on causal discovery and spurious correlation detection. Another area of focus is the application of reinforcement learning to real-world problems, including optimization of infectious disease intervention measures, scheduling in high-performance computing systems, and cooperation evolution mechanisms in spatial public goods games. Noteworthy papers include 'Enhancing Diversity in Parallel Agents: A Maximum State Entropy Exploration Story', which introduces a novel learning framework for parallel reinforcement learning, and 'Mastering Multi-Drone Volleyball through Hierarchical Co-Self-Play Reinforcement Learning', which proposes a hierarchical reinforcement learning framework for multi-drone volleyball. Overall, the field is advancing towards more sophisticated and realistic applications, with a focus on efficiency, scalability, and real-world impact.
Advances in Reinforcement Learning and Multi-Agent Systems
Sources
Enhancing Diversity in Parallel Agents: A Maximum State Entropy Exploration Story
Learning Local Causal World Models with State Space Models and Attention
D3HRL: A Distributed Hierarchical Reinforcement Learning Approach Based on Causal Discovery and Spurious Correlation Detection
Null Counterfactual Factor Interactions for Goal-Conditioned Reinforcement Learning
Multi-Agent Deep Reinforcement Learning for Zonal Ancillary Market Coupling
Unraveling the Rainbow: can value-based methods schedule?
Stochastic scheduling with Bernoulli-type jobs through policy stratification
Multi-Class Stackelberg Games for the Co-Design of Networked Systems
Rainbow Delay Compensation: A Multi-Agent Reinforcement Learning Framework for Mitigating Delayed Observation
Decentralized Distributed Proximal Policy Optimization (DD-PPO) for High Performance Computing Scheduling on Multi-User Systems
Optimization of Infectious Disease Intervention Measures Based on Reinforcement Learning - Empirical analysis based on UK COVID-19 epidemic data
PPO-ACT: Proximal Policy Optimization with Adversarial Curriculum Transfer for Spatial Public Goods Games
Mastering Multi-Drone Volleyball through Hierarchical Co-Self-Play Reinforcement Learning
Adaptive and Robust DBSCAN with Multi-agent Reinforcement Learning
Implicitly Aligning Humans and Autonomous Agents through Shared Task Abstractions
Enhancing Cooperative Multi-Agent Reinforcement Learning with State Modelling and Adversarial Exploration