Utilizing Class Separation Distance for the Evaluation of Corruption Robustness of Machine Learning Classifiers.
Georg SiedelSilvia VockAndrey MorozovStefan VoßPublished in: AISafety@IJCAI (2022)
Keyphrases
- machine learning
- machine learning algorithms
- decision trees
- machine learning methods
- machine learning approaches
- class labels
- supervised classification
- feature selection
- support vector
- learning algorithm
- computer vision
- discriminant functions
- multi class classification
- learning classifier systems
- linear classifiers
- test set
- pattern recognition
- supervised learning
- data sets
- linear discriminant
- learning machines
- training data
- active learning
- learning tasks
- distance measure
- cost sensitive
- inductive logic programming
- text mining
- data mining
- intra class
- multiclass classification
- class probabilities
- multi class problems
- distance function