Improving the Trainability of Deep Neural Networks through Layerwise Batch-Entropy Regularization.
David PeerBart KeulenSebastian StabingerJustus H. PiaterAntonio Jose Rodríguez-SánchezPublished in: Trans. Mach. Learn. Res. (2022)
Keyphrases
- neural network
- pattern recognition
- information theoretic
- information theory
- artificial neural networks
- fuzzy logic
- back propagation
- parameter selection
- prior information
- feed forward
- fuzzy systems
- genetic algorithm
- information entropy
- regularization term
- multilayer perceptron
- neural nets
- self organizing maps
- mutual information
- batch processing