Reinforcement Learning Based Oscillation Dampening: Scaling up Single-Agent RL algorithms to a 100 AV highway field operational test.
Kathy JangNathan LichtléEugene VinitskyAdit ShahMatthew R. BuntingMatthew NiceBenedetto PiccoliBenjamin SeiboldDaniel B. WorkMaria Laura Delle MonacheJonathan SprinkleJonathan W. LeeAlexandre M. BayenPublished in: CoRR (2024)
Keyphrases
- single agent
- rl algorithms
- reinforcement learning
- multi agent
- stochastic games
- multiple agents
- multi agent systems
- model free
- decision problems
- dynamic environments
- optimal control
- actor critic
- function approximation
- state space
- learning algorithm
- cooperative
- learning problems
- path finding
- temporal difference learning
- action space
- markov decision processes
- partially observable
- nash equilibria
- policy gradient
- decision making
- function approximators
- computational complexity
- data mining
- neural network
- learning capabilities
- temporal difference
- adaptive control
- dynamic programming