Selective training of feedforward artificial neural networks using matrix perturbation theory

Neural Networks

Volumn 8, Issue 6, 1995, Pages 931-944

  Hunt, S D ^a   Deller Jr , J R ^b  

^a UNIVERSITY OF PUERTO RICO   (Puerto Rico)

^b MICHIGAN STATE UNIVERSITY   (United States)

Author keywords

Artificial neural networks; Efficient training; Feedforward networks; New training algorithm; Nonlinear optimization; Pattern recognition; Perturbation theory; Selective training; Supervised learning

Indexed keywords

BACKPROPAGATION; ITERATIVE METHODS; LEARNING ALGORITHMS; LEARNING SYSTEMS; MATRIX ALGEBRA; OPTIMIZATION; PATTERN RECOGNITION; PERTURBATION TECHNIQUES;

MATRIX PERTURBATION THEORY; NONLINEAR OPTIMIZATION; SUPERVISED LEARNING;

FEEDFORWARD NEURAL NETWORKS;

ARTICLE; ARTIFICIAL NEURAL NETWORK; FEEDBACK SYSTEM; HUMAN; LEARNING; NONHUMAN; PATTERN RECOGNITION; PRIORITY JOURNAL;

EID: 0028867556     PISSN: 08936080     EISSN: None     Source Type: Journal    
DOI: 10.1016/0893-6080(95)00030-4     Document Type: Article

References (20)

1. Supervised learning process of multilayer perception neural networks using fast least squares
2. An alternative adaptive sequential regression algorithm and its application to the recognition of cerebral palsy speech
3. A simple ‘linearized’ learning algorithm which outperforms backpropagation
4. Learning algorithms for perceptrons using back propagation with selective updates
5. Layer-wise training of feedforward neural networks based on linearization and selective data processing
6. Improving the learning rate of backpropagation with the gradinet reuse algorithm
7. An adaptive least squares algorithm for the efficient training of artificial neural networks
8. An introduction to computing with neural nets
9. Second order backpropagation: implementing an optimal O(n) approximation to Newton's method as an artificial neural network
10. Control methods used in a study of the vowels
11. An artificial neural system design for rotation and scale invariant pattern recognition
12. Determination of the regions of convergence of neural networks using backpropagation and layerwise training algorithms
13. Layered neural nets for pattern recognition