Adapting Loss Functions to Learning Progress Improves Accuracy of Classification in Neural Networks.
Andreas KnoblauchPublished in: ISMIS (2022)
Keyphrases
- loss function
- neural network
- classification accuracy
- partially labeled data
- learning models
- supervised learning
- support vector
- learning tasks
- learning algorithm
- bayes rule
- logistic regression
- pairwise
- pattern classification
- learning problems
- machine learning
- support vector machine
- feature vectors
- feature space
- back propagation
- benchmark datasets
- class labels
- text classification
- multi class
- learning to rank
- training set
- training process
- risk minimization
- decision trees