Memory-based Deep Reinforcement Learning for POMDPs.
Lingheng MengRob GorbetDana KulicPublished in: IROS (2021)
Keyphrases
- reinforcement learning
- partially observable markov decision processes
- partially observable
- function approximation
- state space
- markov decision processes
- policy search
- continuous state
- optimal policy
- multi agent
- dynamic programming
- control problems
- reinforcement learning algorithms
- temporal difference
- model free
- hidden state
- policy gradient
- memory based learning
- machine learning
- learning problems
- learning algorithm
- policy iteration algorithm
- optimal control
- neural network
- continuous state spaces
- actor critic
- learning process
- reinforcement learning methods
- policy iteration
- transfer learning
- partial observability
- linear programming
- mobile robot
- action space
- learning agent
- markov decision process
- point based value iteration
- single agent