What size network is good for generalization of a specific task of interest?

Neural Networks

Volumn 7, Issue 2, 1994, Pages 321-329

  Amirikian, Bagrat ^a,b   Nishimura, Hajime ^a  

^a UNIVERSITY OF TOKYO   (Japan)

^b RUSSIAN ACADEMY OF SCIENCES   (Russian Federation)

Author keywords

Boolean function; Capacity; Complexity; Feedforward; Generalization; Learning; Neural network; Occam's razor; Training

Indexed keywords

BINARY SEQUENCES; BOOLEAN FUNCTIONS; COMPUTATIONAL COMPLEXITY; COMPUTER ARCHITECTURE; FORMAL LOGIC; LEARNING SYSTEMS; NETWORK PROTOCOLS; PATTERN RECOGNITION; PROBABILITY; RANDOM PROCESSES; SYSTEMS ANALYSIS; THRESHOLD ELEMENTS;

BOOLEAN CLASSIFICATION TASKS; FEEDFORWARD NEURAL NETWORKS; GENERALIZATION; INFORMATION ENTROPY; NEURAL NETWORK SIZE; NEURAL NETWORK TRAINING; OCCAM'S RAZOR;

NEURAL NETWORKS;

EID: 0028195682     PISSN: 08936080     EISSN: None     Source Type: Journal    
DOI: 10.1016/0893-6080(94)90026-4     Document Type: Article

References (22)

1. A universal theorem on learning curves
2. Four types of learning curves
3. A neural network learns trajectory of motion from the least action principle
4. On the ability of neural networks to perform generalization by induction
5. What size net gives valid generalization?
6. Learnability and the Vapnik-Chervonenkis dimension
7. Exhaustive thermodynamical analysis of Boolean learning networks
8. How tight are the Vapnik-Chervonenkis bounds?
9. Geometrical and statistical properties of systems of linear inequalities with applications in pattern recognition
10. Large automatic learning, rule extraction, and generalization
11. A general lower bound on the number of examples needed for learning
12. A statistical approach to learning and generalization in layered neural networks
13. Learning networks of neurons with Boolean logic
14. On the connection between the complexity and credibility of inferred models
15. Exhaustive Learning
16. A theory of the learnable
17. Learning in feedforward Boolean networks
18. On the uniform convergence of relative frequences of events to their probabilities
19. The relationship between Occam's razor and convergent guessing
20. Stacked generalization