Reaching Pruning Locations in a Vine Using a Deep Reinforcement Learning Policy.
Francisco YandúnTanvir ParharAbhisesh SilwalDavid CliffordZhiqiang YuanGabriella LevineSergey YaroshenkoGeorge KantorPublished in: ICRA (2021)
Keyphrases
- reinforcement learning
- optimal policy
- policy search
- markov decision process
- action selection
- policy iteration
- state space
- function approximators
- partially observable environments
- partially observable
- reward function
- reinforcement learning problems
- pruning method
- policy gradient
- actor critic
- markov decision processes
- state and action spaces
- action space
- policy evaluation
- function approximation
- reinforcement learning algorithms
- search space
- model free
- markov decision problems
- infinite horizon
- control policy
- approximate dynamic programming
- state action
- dynamic programming
- decision problems
- rl algorithms
- partially observable markov decision processes
- continuous state spaces
- learning problems
- temporal difference
- machine learning
- control policies
- model free reinforcement learning
- policy gradient methods
- temporal difference learning
- average reward
- long run
- learning algorithm
- finite state
- pruning algorithm
- pruning methods
- deep learning
- asymptotically optimal
- average cost
- transfer learning
- partially observable domains
- learning process