Continuous-action Reinforcement Learning for Playing Racing Games: Comparing SPG to PPO.
Mario S. HolubarMarco A. WieringPublished in: CoRR (2020)
Keyphrases
- continuous action
- reinforcement learning
- policy search
- continuous state
- computer games
- game playing
- action space
- reinforcement learning algorithms
- learning agents
- imperfect information
- state space
- partially observable markov decision processes
- learning algorithm
- finite state
- action selection
- function approximation
- video games
- robot navigation
- multi agent
- model free
- game theory
- dynamic programming
- nash equilibria
- state dependent
- stochastic games
- optimal policy
- control policies
- markov decision problems
- multi agent systems
- search algorithm