Multi-Step Recurrent Q-Learning for Robotic Velcro Peeling.
Jiacheng YuanNicolai HäniVolkan IslerPublished in: CoRR (2020)
Keyphrases
- multi step
- td learning
- reinforcement learning
- function approximation
- cooperative
- lower bounding
- multi agent
- robotic systems
- recurrent neural networks
- learning algorithm
- mobile robot
- state space
- model free
- action selection
- single step
- real robot
- knn
- k nearest neighbor
- semi supervised
- data sets
- reinforcement learning algorithms
- policy iteration
- optimal policy
- dimensionality reduction
- distance computation
- input data