A Multi-armed Bandit to Smartly Select a Training Set from Big Medical Data.
Benjamín GutiérrezLoïc PeterTassilo KleinChristian WachingerPublished in: MICCAI (3) (2017)
Keyphrases
- medical data
- training set
- multi armed bandit
- medical images
- multi armed bandits
- clinical data
- medical data mining
- medical information
- training data
- data sets
- supervised learning
- data mining
- medical records
- classification accuracy
- feature space
- patient outcomes
- classification algorithm
- active learning
- reinforcement learning
- patient care
- decentralized decision making
- privacy sensitive
- patient data
- training samples
- support vector machine
- decision trees
- regret bounds
- class labels
- pairwise