A Novel Information-Theoretic Objective to Disentangle Representations for Fair Classification.
Pierre ColomboNathan NoiryGuillaume StaermanPablo PiantanidaPublished in: CoRR (2023)
Keyphrases
- information theoretic
- information theory
- mutual information
- theoretic framework
- entropy measure
- log likelihood
- classification accuracy
- pattern recognition
- information bottleneck
- jensen shannon divergence
- information theoretic measures
- relative entropy
- decision trees
- minimum description length
- feature extraction
- kullback leibler divergence
- supervised learning
- text classification
- support vector
- multi modality
- computational learning theory
- jensen shannon
- density estimation
- kl divergence
- data mining
- training set
- feature selection