Scaling All-Goals Updates in Reinforcement Learning Using Convolutional Neural Networks.
Fabio PardoVitaly LevdikPetar KormushevPublished in: AAAI (2020)
Keyphrases
- convolutional neural networks
- reinforcement learning
- convolutional network
- function approximation
- model free
- optimal policy
- state space
- reinforcement learning algorithms
- temporal difference learning
- learning algorithm
- machine learning
- transfer learning
- data sets
- action selection
- policy search
- function approximators
- neural network
- data structure
- robotic control
- multi agent reinforcement learning
- reinforcement learning methods
- action space
- partially observable
- optimal control
- supervised learning
- multiresolution