Information-theoretic multi-layered supervised self-organizing maps for improved prediction performance and explicit internal representation.
Ryotaro KamimuraPublished in: SMC (2014)
Keyphrases
- information theoretic
- multi layered
- self organizing maps
- competitive learning
- mutual information
- information theory
- unsupervised learning
- theoretic framework
- neural network
- information bottleneck
- jensen shannon divergence
- neural gas
- input data
- network anomaly detection
- log likelihood
- supervised learning
- kullback leibler divergence
- minimum description length
- information theoretic measures
- entropy measure
- learning algorithm
- image classification
- semi supervised
- kl divergence
- decision trees
- feature selection
- expectation maximization
- k means
- relative entropy
- feature extraction