A generalized meta-loss function for distillation and learning using privileged information for classification and regression.
Amina AsifMuhammad DawoodFayyaz ul Amir Afsar MinhasPublished in: CoRR (2018)
Keyphrases
- loss function
- bayes rule
- support vector
- supervised learning
- privileged information
- boosting framework
- learning algorithm
- active learning
- classification accuracy
- stochastic gradient descent
- histogram intersection kernel
- generalization ability
- learning tasks
- training samples
- logistic regression
- text classification
- gradient boosting
- hinge loss
- support vector machine
- pairwise
- feature space
- feature extraction