Deep Reinforcement Learning in VizDoom via DQN and Actor-Critic Agents.
Maria BakhanovaIlya MakarovPublished in: IWANN (1) (2021)
Keyphrases
- actor critic
- reinforcement learning
- multi agent
- temporal difference
- policy gradient
- reinforcement learning algorithms
- action selection
- single agent
- optimal control
- function approximation
- multi agent systems
- gradient method
- approximate dynamic programming
- learning agent
- neuro fuzzy
- multiple agents
- policy iteration
- model free
- cooperative
- state space
- dynamic environments
- stochastic games
- decision making
- learning capabilities
- control problems
- markov decision processes
- optimal policy
- policy gradient methods
- partially observable markov decision processes
- incomplete information
- reinforcement learning methods
- rl algorithms
- markov decision process
- action space
- average reward
- transfer learning
- supervised learning
- mobile robot
- machine learning