Adjoint operators enable fast and amortized machine learning based Bayesian uncertainty quantification.
Rafael OrozcoAli SiahkoohiGabrio RizzutiTristan van LeeuwenFelix J. HerrmannPublished in: Medical Imaging: Image Processing (2023)
Keyphrases
- machine learning
- decision theory
- dempster shafer
- pattern recognition
- worst case
- bayesian methods
- bayesian inference
- bayesian networks
- inductive learning
- machine learning algorithms
- computer vision
- handling uncertainty
- statistical inference
- explanation based learning
- decision trees
- machine learning methods
- natural language processing
- data mining
- optimal control
- knowledge discovery
- learning systems
- search tree
- maximum likelihood
- active learning
- gaussian processes
- morphological operators
- covariate shift
- uncertain data
- statistical methods
- knowledge acquisition
- text classification
- artificial intelligence
- reinforcement learning
- conditional probabilities
- data analysis
- knowledge representation
- learning algorithm
- uncertain information
- expected utility
- query processing
- probability distribution
- support vector machine
- information extraction
- computational intelligence
- learning tasks