Boosting Randomized Smoothing with Variance Reduced Classifiers.
Miklós Z. HorváthMark Niklas MüllerMarc FischerMartin T. VechevPublished in: ICLR (2022)
Keyphrases
- ensemble learning
- weak classifiers
- randomized trees
- decision stumps
- ensemble classifier
- weak learners
- decision forest
- improving classification accuracy
- feature selection
- variance reduction
- boosting framework
- bias variance decomposition
- adaboost algorithm
- boosting algorithms
- majority voting
- support vector
- strong classifier
- decision trees
- accurate classifiers
- multiple classifier systems
- ensemble methods
- ensemble classification
- training samples
- training set
- linear classifiers
- binary classification problems
- classification systems
- concept drift
- test set
- combining classifiers
- naive bayes classifier
- meta learning
- training data
- naive bayes
- weighted voting
- learning algorithm
- bayesian classifiers
- multiclass classification
- classifier ensemble
- multi class
- roc curve
- random forests
- base classifiers
- loss function
- base learners
- classification trees
- active learning
- feature set
- covariance matrix