A5077477958- Reinforcement Learning in Robotics
- Artificial Intelligence in Games
- Access Control and Trust
- High voltage insulation and dielectric phenomena
- Guidance and Control Systems
- Advanced Sensor and Energy Harvesting Materials
- Advanced Bandit Algorithms Research
- Simulation Techniques and Applications
- Dielectric materials and actuators
Beijing University of Posts and Telecommunications
2022-2025
Self-play methods have achieved remarkable success in two-player zero-sum games, attaining superhuman performance many complex game domains. Parallelizing learners is a feasible approach to handle games. However, parallelizing often leads the suboptimal exploitation of computational resources, resulting inefficiencies. This paper introduces Mixed Hierarchical Oracle (MHO), which designed enhance training efficiency and MHO efficiently leverages interaction data among parallelized solvers...
Policy space response oracles (PSRO) is an important algorithmic framework for approximating Nash equilibria in two-player zero-sum games. Enhancing policy diversity has been shown to improve the performance of PSRO this approximation process significantly. However, existing metrics are often prone redundancy, which can hinder optimal strategy convergence. In paper, we introduce similarity measure (PSM), a novel approach that combines Gaussian and cosine measures assess similarity. We...
Policy space response oracle (PSRO) is a population-based algorithm that can be used to solve two-player zero-sum games. In the PSRO solution framework, optimizing policy diversity crucial for addressing nontransitive game problems, helping agent population avoid exploitation by unfamiliar opponents. addition, while deep reinforcement learning highly effective in solving complex environments, its integration with remains fragmented and lacking coordination. this study, we propose distributed...
Polymer-based dielectrics with high energy storage density are attracting increasing attention due to their wide applications in pulsed-discharge and power conditioning electronic fields. Despite some numerical simulation about effects of horizontally arranged vertically fibers on dielectric properties composites already studied, the influence mechanism specific orientation aspect ratio still remains be studied. In this work, angles ratios nanofiber fillers breakdown behavior theoretically...