A two-head loss function for deep Average-K classification.
Camille GarcinMaximilien ServajeanAlexis JolyJoseph SalmonPublished in: CoRR (2023)
Keyphrases
- loss function
- support vector
- hinge loss
- pairwise
- classification accuracy
- learning to rank
- pattern classification
- logistic regression
- risk minimization
- bayes rule
- decision trees
- reproducing kernel hilbert space
- boosting framework
- empirical risk
- training samples
- support vector machine
- feature selection
- text classification
- regularization term
- feature vectors
- machine learning algorithms
- classification algorithm
- image classification
- euclidean space
- generalization ability
- similarity measure
- feature extraction
- machine learning