Reinforcement learning in many-agent settings under partial observability.
Keyang HePrashant DoshiBikramjit BanerjeePublished in: UAI (2022)
Keyphrases
- partial observability
- reinforcement learning
- partially observable
- learning agent
- agent programming
- belief state
- markov decision process
- state space
- symbolic model checking
- fully observable
- markov decision processes
- multi agent
- action selection
- partial information
- reward function
- belief space
- multi agent systems
- learning capabilities
- multiagent systems
- function approximation
- single agent
- reinforcement learning algorithms
- decision problems
- solving problems
- partially observable markov decision processes
- multiple agents
- planning under partial observability
- planning problems
- machine learning
- autonomous agents
- optimal policy
- learning algorithm
- learning tasks
- markov decision problems
- dynamical systems
- hidden state
- decision making
- decision theoretic
- transfer learning