SafeML: Safety Monitoring of Machine Learning Classifiers through Statistical Difference Measure.
Koorosh AslansefatIoannis SorokosDeclan WhitingRamin Tavakoli KolagariYiannis PapadopoulosPublished in: CoRR (2020)
Keyphrases
- machine learning
- machine learning algorithms
- machine learning methods
- machine learning approaches
- statistical methods
- feature selection
- decision trees
- diversity measures
- supervised classification
- confusion matrix
- information theory
- statistical measures
- multiple classifier systems
- real time
- statistical tests
- statistical analysis
- distance measure
- test set
- statistical inference
- training set
- pattern recognition
- learning tasks
- statistical models
- monitoring system
- learning classifier systems
- classification models
- training data
- learning algorithm
- data mining
- inductive logic programming
- information theoretic
- ensemble learning
- knowledge acquisition
- supervised learning
- support vector
- nuclear power plant
- similarity measure
- correlation analysis
- normal operation
- training samples
- classification algorithm
- svm classifier
- training examples
- semi supervised learning
- text classification
- feature set
- text mining
- natural language processing
- support vector machine
- data analysis
- computer science
- reinforcement learning
- computer vision
- artificial intelligence
- data sets