Genetic Weight Optimization of a Feedforward Neural Network Controller
作者： Dirk Thierens , Johan Suykens , Joos Vandewalle , Bart De Moor
DOI: 10.1007/978-3-7091-7533-0_95
关键词: Control theory 、 Stochastic neural network 、 Time delay neural network 、 Physical neural network 、 Intelligent control 、 Equilibrium point 、 Machine learning 、 Probabilistic neural network 、 Nonlinear system 、 Feedforward neural network 、 Artificial intelligence 、 Inhibitory postsynaptic potential 、 Control theory 、 Artificial neural network 、 Excitatory postsynaptic potential 、 Recurrent neural network 、 Computer science
摘要: The optimization of the weights a feedforward neural network with genetic algorithm is discussed. search by recombination operator hampered existence two functional equivalent symmetries in networks. To sidestep these representation redundancies we reorder hidden neurons on genotype before according to weight sign matching criterion, and flip signs neuron’s connections whenever there are more inhibitory than excitatory incoming outgoing links. As an example optimize that implements nonlinear optimal control law. controller has swing up inverted pendulum from its lower equilibrium point upper stabilize it there. Finding represents problem which solved algorithm.
springer.com
core.ac.uk
sci-hub.st 参考文章(9)
David E. Goldberg, Kalyanmoy Deb, An Investigation of Niche and Species Formation in Genetic Function Optimization international conference on genetic algorithms. pp. 42- 50 ,(1989)
D Whitley, T Starkweather, C Bogart, Genetic algorithms and neural networks: optimizing connections and connectivity parallel computing. ,vol. 14, pp. 347- 361 ,(1990) , 10.1016/0167-8191(90)90086-O
Ken-Ichi Funahashi, On the approximate realization of continuous mappings by neural networks Neural Networks. ,vol. 2, pp. 183- 192 ,(1989) , 10.1016/0893-6080(89)90003-8
Héctor J. Sussmann, Original Contribution: Uniqueness of the weights for minimal feedforward nets with a given input-output map Neural Networks. ,vol. 5, pp. 589- 593 ,(1992) , 10.1016/S0893-6080(05)80037-1
Halbert White, Connectionist nonparametric regression: multilayer feedforward networks can learn arbitrary mappings Neural Networks. ,vol. 3, pp. 535- 549 ,(1990) , 10.1016/0893-6080(90)90004-5
Kurt Hornik, Maxwell Stinchcombe, Halbert White, Multilayer feedforward networks are universal approximators Neural Networks. ,vol. 2, pp. 359- 366 ,(1989) , 10.1016/0893-6080(89)90020-8
Hans-Paul Schwefel, Numerical optimization of computer models ,(1981)
Arthur E. Bryson, Applied optimal control ,(1969)
HornikK., StinchcombeM., WhiteH., Multilayer feedforward networks are universal approximators Neural Networks. ,(1989) , 10.5555/70405.70408