Instance-based Label Smoothing For Better Calibrated Classification Networks.
Mohamed MaherMeelis KullPublished in: ICMLA (2021)
Keyphrases
- class labels
- classification accuracy
- pattern recognition
- pattern classification
- classification algorithm
- instance based learning
- support vector machine svm
- classification scheme
- automatic classification
- label ranking
- feature selection
- feature extraction
- classification method
- machine learning methods
- unsupervised learning
- knn
- training set
- classification process
- bayesian networks
- decision trees
- classification models
- cost sensitive
- classification systems
- machine learning
- multi view
- image classification
- support vector machine
- feature space
- multi label
- decision rules
- training samples
- multi class
- feature vectors
- heterogeneous networks
- support vector
- data sets