Contextual Reinforcement Learning of Visuo-tactile Multi-fingered Grasping Policies.
Visak KumarTucker HermansDieter FoxStan BirchfieldJonathan TremblayPublished in: CoRR (2019)
Keyphrases
- reinforcement learning
- object manipulation
- optimal policy
- control theoretic
- robot control
- policy search
- markov decision process
- function approximation
- reward function
- reinforcement learning agents
- hierarchical reinforcement learning
- multiagent reinforcement learning
- control policies
- control policy
- fitted q iteration
- model free
- learning algorithm
- visual feedback
- reinforcement learning algorithms
- context dependent
- context sensitive
- markov decision processes
- sufficient conditions
- machine learning