A Variational EM Approach to Predicting Uncertainty in Supervised Learning.
Markus HarvaPublished in: IJCNN (2006)
Keyphrases
- supervised learning
- unsupervised learning
- image segmentation
- expectation maximization
- em algorithm
- training data
- machine learning
- statistical learning
- active learning
- semi supervised
- generative model
- learning algorithm
- class labels
- uncertain data
- mixture model
- maximum likelihood
- decision theory
- training set
- reinforcement learning
- computer vision
- supervised classification
- supervised machine learning
- learning problems
- learning tasks
- image denoising
- gaussian mixture model
- bayesian framework
- training examples
- parameter estimation
- semi supervised learning
- multiple instance learning
- belief functions
- level set
- uncertain information
- free energy
- optical flow computation
- image processing