Sufficiently Informative and Relevant Features: An Information-Theoretic and Fourier-Based Characterization.
Mohsen HeidariJithin K. SreedharanGil I. ShamirWojciech SzpankowskiPublished in: IEEE Trans. Inf. Theory (2022)
Keyphrases
- information theoretic
- mutual information
- information theory
- theoretic framework
- highly informative
- entropy measure
- jensen shannon divergence
- information theoretic measures
- information bottleneck
- multi modality
- log likelihood
- kullback leibler divergence
- relative entropy
- feature set
- prior knowledge
- feature vectors
- feature extraction
- data mining
- distributional clustering
- computational learning theory
- kl divergence
- image features
- feature selection