Improving Reinforcement Learning Pre-Training with Variational Dropout.
Tom BlauLionel OttFabio RamosPublished in: IROS (2018)
Keyphrases
- reinforcement learning
- supervised learning
- image segmentation
- training set
- learning algorithm
- multi agent
- dynamic programming
- neural network
- function approximation
- machine learning
- model free
- training phase
- training algorithm
- training process
- test set
- state space
- learning process
- optical flow computation
- reinforcement learning methods
- multi agent reinforcement learning
- free energy
- temporal difference
- markov decision processes
- optimal policy
- support vector machine