Minimum class entropy: A maximum information approach to layered networks

Neural Networks

Volumn 2, Issue 2, 1989, Pages 133-141

  Bichsel, Martin ^b   Seitz, Peter ^a  

^a PAUL SCHERRER INSTITUTE   (Switzerland)

^b ETH ZURICH   (Switzerland)

Author keywords

Information theory; Layered networks; Minimum entropy; Performance measure; Teaching algorithms

Indexed keywords

CODES, SYMBOLIC--ENCODING; COMPUTER PROGRAMMING--ALGORITHMS; INFORMATION THEORY;

FEEDFORWARD NETWORKS; LAYERED NETWORKS; MINIMUM CLASS ENTROPY; MULTISTAGE ENCODERS; TEACHING ALGORITHMS; TRAINING ALGORITHMS;

SYSTEMS SCIENCE AND CYBERNETICS;

EID: 0024775620     PISSN: 08936080     EISSN: None     Source Type: Journal    
DOI: 10.1016/0893-6080(89)90030-0     Document Type: Article

References (13)

1. Minimizing multimodal functions of continuous variables with the “simulated annealing” algorithm
2. Application of neural network algorithms and architectures to correlation/tracking and identification
3. Optimization by simulated annealing
4. Self-organization in a perceptual neural network
5. An introduction to computing with neural nets
6. A logical calculus of the ideas immanent in nervous activity
7. Learning symmetry groups with hidden units: Beyond the perceptron
8. A mathematical theory of communication
9. A mathematical theory of communication
10. A Monte Carlo simulated annealing approach to optimization over continuous variables