Minimax Classification under Concept Drift with Multidimensional Adaptation and Performance Guarantees.
Verónica ÁlvarezSantiago MazuelasJosé Antonio LozanoPublished in: ICML (2022)
Keyphrases
- concept drift
- data stream classification
- classification algorithm
- data streams
- ensemble classifier
- class imbalance
- novelty detection
- non stationary
- concept drifting data streams
- decision trees
- pattern recognition
- machine learning
- labelled data
- drifting concepts
- support vector machine
- change detection
- incremental learning
- text classification
- training set
- data stream mining
- data sets
- neyman pearson
- data distribution
- classification accuracy
- feature vectors
- feature extraction
- evolving data streams
- feature space
- incremental clustering
- concept descriptions
- small number
- class labels
- benchmark datasets
- dimensionality reduction
- unsupervised learning