Adaptable Automation with Modular Deep Reinforcement Learning and Policy Transfer.
Zohreh RazieiMohsen MoghaddamPublished in: CoRR (2020)
Keyphrases
- reinforcement learning
- optimal policy
- policy search
- transfer learning
- action selection
- markov decision process
- function approximators
- partially observable environments
- markov decision processes
- actor critic
- reinforcement learning problems
- partially observable
- control policy
- policy iteration
- reinforcement learning algorithms
- state space
- control policies
- reward function
- state and action spaces
- approximate dynamic programming
- policy evaluation
- markov decision problems
- long run
- model free
- policy gradient
- decision problems
- function approximation
- continuous state spaces
- average reward
- model free reinforcement learning
- dynamic programming
- learning problems
- knowledge transfer
- action space
- partially observable markov decision processes
- temporal difference
- state action
- infinite horizon
- rl algorithms
- continuous state
- reinforcement learning methods
- state dependent
- temporal difference learning
- modular structure
- learning process
- multi agent systems
- agent receives
- learning algorithm
- partially observable domains
- agent learns
- learning tasks
- optimal control