SOAP-RL: Sequential Option Advantage Propagation for Reinforcement Learning in POMDP Environments.
Shu IshidaJoão F. HenriquesPublished in: CoRR (2024)
Keyphrases
- reinforcement learning
- continuous state
- function approximation
- reinforcement learning algorithms
- state space
- web services
- hidden state
- markov decision processes
- partially observable markov decision processes
- model free
- partially observable
- learning algorithm
- temporal difference
- markov decision process
- model free reinforcement learning
- transfer learning
- multi agent environments
- dynamic environments
- description language
- multi agent
- rl algorithms
- machine learning
- reward function
- control problems
- learning process
- markov decision problems
- optimal control
- policy evaluation
- partial observability
- reinforcement learning methods
- function approximators
- learning problems
- action space
- policy iteration
- dynamic programming
- markov chain
- optimal policy
- action selection
- direct policy search