Reinforcement learning with model-based feedforward inputs for robotic table tennis.
Hao MaDieter BüchlerBernhard SchölkopfMichael MuehlebachPublished in: Auton. Robots (2023)
Keyphrases
- feed forward
- table tennis
- reinforcement learning
- model free
- back propagation
- key features
- artificial neural networks
- neural network
- neural nets
- hidden layer
- output layer
- biologically plausible
- visual cortex
- mobile robot
- neural architecture
- real robot
- real time
- state space
- adaptive neural
- recurrent neural networks
- dynamic programming
- machine learning
- skill learning
- action selection
- recurrent networks
- markov decision processes
- genetic algorithm
- action space
- multiscale
- optimal policy