Improving Classifier Confidence using Lossy Label-Invariant Transformations.
Sooyong JangInsup LeeJames WeimerPublished in: CoRR (2020)
Keyphrases
- class labels
- label noise
- high confidence
- multi label classification
- preserving transformations
- quasi invariant
- similarity transformations
- support vector machine
- learning algorithm
- confidence estimates
- semantic labels
- image transformations
- geometric transformations
- training data
- classification algorithm
- support vector
- svm classifier
- feature selection
- affine transformation
- multilabel classification
- feature space
- feature set
- data sets
- decision trees
- class membership
- confidence level
- training set
- learning classifier systems
- linear classifiers
- classification rate
- classification scheme
- multi label
- classification method
- nearest neighbor
- invariant features
- confidence scores
- data compression
- feature extraction
- training samples