Improving Spatiotemporal Self-Supervision by Deep Reinforcement Learning.
Uta BüchlerBiagio BrattoliBjörn OmmerPublished in: CoRR (2018)
Keyphrases
- reinforcement learning
- space time
- multi agent
- function approximation
- moving objects
- optimal control
- optimal policy
- spatiotemporal data
- stochastic approximation
- temporal difference
- model free
- spatial and temporal
- markov decision processes
- databases
- sufficient conditions
- evolutionary algorithm
- artificial neural networks
- website
- machine learning