Error sensitivity analysis of Delta divergence - a novel measure for classifier incongruence detection.
Josef KittlerCemre ZorIoannis KaloskampisYulia HicksWenwu WangPublished in: Pattern Recognit. (2018)
Keyphrases
- sensitivity analysis
- divergence measure
- managerial insights
- kullback leibler
- false negative
- support vector machine classifier
- influence diagrams
- object detection
- distance measure
- false positives
- detection algorithm
- variational inequalities
- receiver operating characteristic curves
- similarity measure
- training data
- confusion matrix
- mutual information
- neyman pearson
- detection method
- feature space
- training set
- error rate
- special case
- error detection
- error measure
- kl divergence
- training samples
- learning algorithm
- decision making
- support vector machine
- jensen shannon divergence
- minimum error
- multi view
- detection rate
- relative entropy
- classification algorithm
- object detectors