Download:
|
by Byoung-tak Zhang, Peter Ohm, Heinz Muhlenbein
Evolutionary Computation
http://comp.snu.ac.kr/~btzhang/ai-lab/publications/ecj97.ps.gz
Add To MetaCart
Abstract:
This paper is concerned with the automatic induction of parsimonious neural networks. In contrast to other program induction situations, network induction entails parametric learning as well as structural adaptation. We present a novel representation scheme, called neural trees, that allows efficient learning of network architectures and parameters both by genetic search. A hybrid evolutionary method is developed for neural tree induction that combines genetic programming and the breeder genetic algorithm under the unified framework of the minimum description length principle. The method is successfully applied to the induction of higher-order neural trees, while still keeping the resulting structures sparse to ensure good generalization performance. Empirical results are provided on two chaotic time-series prediction problems of practical interest.
Citations
|
1782
|
Genetic Programming: On the Programming of Computers by Means of Natural Selection Cambridge
– Koza
- 1992
|
|
489
|
Neural networks and the bias/variance dilemma
– Geman, Bienenstock, et al.
- 1992
|
|
250
|
Designing neural network using genetic algorithm with graph generation system
– Kitano
- 1990
|
|
181
|
Evolving networks: Using the genetic algorithm with connectionist learning
– Belew, McInerney, et al.
- 1992
|
|
159
|
An evolutionary algorithm that constructs recurrent neural networks
– Angeline, Saunders, et al.
- 1993
|
|
94
|
The science of breeding and its application to the breeder genetic algorithm
– Muhlenbein, Schlierkamp-Voosen
- 1994
|
|
92
|
Perceptrons
– Minsky, Papert
- 1969
|
|
76
|
Coevolving high-level representations
– Angeline, Pollack
- 1994
|
|
59
|
Balancing accuracy and parsimony in genetic programming
– Zhang, Mühlenbein
- 1995
|
|
47
|
Learning, invariance, and generalization in highorder neural networks
– Giles, Maxwell
- 1987
|
|
45
|
Evolutionary design of neural architectures
– Balakrishnan, Honavar
- 1995
|
|
39
|
Genetic synthesis of modular neural networks
– Gruau
- 1993
|
|
37
|
Evolving neural networks
– Fogel, Fogel, et al.
- 1990
|
|
22
|
Extending genetic programming with recombinative guidance
– Iba, Garis
- 1996
|
|
22
|
Stochastic complexity and modeling. The Annals of Statistics
– Rissanen
- 1986
|
|
18
|
System identification using structured genetic algorithms
– Iba, Kurita, et al.
- 1993
|
|
13
|
Genetic Algorithms and Walsh Functions
– Goldberg
- 1989
|
|
13
|
Simultaneous discovery of reusable detectors and subroutines using genetic programming
– Koza
- 1993
|
|
11
|
Towards automatic discovery of building blocks in genetic programming
– Rosca
- 1995
|
|
7
|
Lorenz-Like Chaos in NH3-FIR Lasers
– Hübner, Weiss, et al.
- 1993
|
|
7
|
Developing higher-order networks with empirically selected units
– Kowalczyk, Ferr'a
- 1994
|
|
7
|
Synthesis of sigma-pi neural networks by the breeder genetic programming
– Muhlenbein, Zhang
- 1994
|
|
6
|
Structural adaptation of parsimonious higher-order neural classifiers. Neural Networks
– Fahner, Eckmiller
- 1994
|
|
3
|
Investigating the generality of automatically defined functions, in 'Proceedings of the first annual conference on genetic programming
– O'Reilly
- 1996
|
|
2
|
Polynomial theory of complex systems
– Ivanknenko
- 1971
|
