1 code implementation • 24 Jul 2020 • Jianyu Su, Stephen Adams, Peter A. Beling
To obtain a reasonable trade-off between training efficiency and algorithm performance, we extend value-decomposition to actor-critics that are compatible with A2C and propose a novel actor-critic framework, value-decomposition actor-critics (VDACs).
Multi-agent Reinforcement Learning Reinforcement Learning (RL) +2
no code implementations • 1 Apr 2020 • Jianyu Su, Stephen Adams, Peter A. Beling
The flexibility of the graph structure enables our method to be applicable to a variety of multi-agent systems, e. g. dynamic systems that consist of varying numbers of agents and static systems with a fixed number of agents.
no code implementations • 22 Nov 2019 • Jianyu Su, Peter A. Beling, Rui Guo, Kyungtae Han
The ability to model and predict ego-vehicle's surrounding traffic is crucial for autonomous pilots and intelligent driver-assistance systems.