Rearrangement with Nonprehensile Manipulation Using Deep Reinforcement Learning.
Weihao YuanJohannes A. StorkDanica KragicMichael Yu WangKaiyu HangPublished in: CoRR (2018)
Keyphrases
- reinforcement learning
- function approximation
- learning algorithm
- state space
- reinforcement learning algorithms
- model free
- temporal difference
- optimal policy
- dynamic programming
- markov decision processes
- multi agent
- stochastic approximation
- information retrieval
- robotic control
- direct policy search
- deep learning
- learning problems
- partially observable
- temporal difference learning
- optimal control
- data sets
- machine learning
- neural network
- learning classifier systems
- sufficient conditions
- mobile robot
- action space
- function approximators
- decision trees
- information systems
- autonomous learning