VASE: Variational Assorted Surprise Exploration for Reinforcement Learning.
Haitao XuBrendan McCaneLech SzymanskiPublished in: CoRR (2019)
Keyphrases
- reinforcement learning
- active exploration
- exploration strategy
- action selection
- exploration exploitation
- model based reinforcement learning
- image segmentation
- exploration exploitation tradeoff
- autonomous learning
- reinforcement learning algorithms
- function approximation
- markov decision processes
- state space
- optical flow
- learning algorithm
- multi agent
- optimal policy
- data sets
- balancing exploration and exploitation
- machine learning
- free energy
- temporal difference
- model free
- learning problems
- information visualization
- partially observable
- variational methods
- optical flow computation
- unknown environments
- evolutionary algorithm
- interactive exploration
- multiscale
- neural network
- least squares