Generalization properties of modular networks: Implementing the parity function

IEEE Transactions on Neural Networks

Volumn 12, Issue 6, 2001, Pages 1306-1313

  Franco, Leonardo ^a,b   Cannas, Sergio Alejandro ^a  

^a UNIVERSIDAD NACIONAL DE CÓRDOBA   (Argentina)

^b SISSA   (Italy)

Author keywords

Circuit complexity; Generalization; Learning from examples; Modular neural networks; Parity function

Indexed keywords

BOOLEAN FUNCTIONS; COMPUTATIONAL COMPLEXITY; COMPUTER SIMULATION; LEARNING ALGORITHMS;

CIRCUIT COMPLEXITY; GENERALIZATION ABILITY; MODULAR NEURAL NETWORKS; PARITY FUNCTION;

NEURAL NETWORKS;

EID: 0035505551     PISSN: 10459227     EISSN: None     Source Type: Journal    
DOI: 10.1109/72.963767     Document Type: Article

References (18)

1. Multiexpert and hybrid connectionist approach for pattern recognition: Speaker identification task
2. Exhaustive thermodynamical analysis of Boolean learning networks
3. A self-generating modular neural-network architecture for supervised learning
4. Organization of face and object recognition in modular neural-network models
5. Large automatic learning, rule extraction and generalization
6. Solving arithmetic problems using feed-forward neural networks
7. Generalization and selection of examples in feedforward neural networks
8. Minimal feedforward parity networks using threshold gates
9. Size-depth tradeoffs for threshold circuits
10. Circuit complexity and feedforward neural networks
11. Learning networks of neutrons with boolean logic
12. Scaling relationships in backpropagation learning: Dependence on predicate order
13. Learning in feedforward Boolean networks