Out-Of-Distribution As A Target Class in Semi-Supervised Learning.
Antoine TadrosSébastien DrouyerRafael Grompone von GioiPublished in: ICASSP (2022)
Keyphrases
- semi supervised learning
- target class
- labeled data
- labeled examples
- semi supervised
- unlabeled data
- transfer learning
- co training
- supervised learning
- data distribution
- text categorization
- training data
- machine learning
- unsupervised learning
- active learning
- learning problems
- text classification
- class distribution
- multiple instance learning
- training instances
- learning algorithm
- class probabilities
- support vector machine
- class labels
- target object
- learning experience
- object class
- bounding box
- training set
- data sets
- reinforcement learning
- data streams
- data points
- pairwise