Method and system for training a neural network with adaptive we

Details Method and system for training a neural network with adaptive we Method and system for training a neural network with adaptive we

1997-04-29

1998-09-22

Hafiz, Tariq R.

Data processing: artificial intelligence

Neural network

Learning method

706 15, 706 16, 706 21, G06F 1518

Patent

active

058129928

ABSTRACT:
A signal processing system and method for accomplishing signal processing using a neural network that incorporates adaptive weight updating and adaptive pruning for tracking non-stationary signal is presented. The method updates the structural parameters of the neural network in principal component space (eigenspace) for every new available input sample. The non-stationary signal is recursively transformed into a matrix of eigenvectors with a corresponding matrix of eigenvalues. The method applies principal component pruning consisting of deleting the eigenmodes corresponding to the smallest saliencies, where a sum of the smallest saliencies is less than a predefined threshold level. Removing eigenmodes with low saliencies reduces the effective number of parameters and generally improves generalization. The output is then computed by using the remaining eigenmodes and the weights of the neural network are updated using adaptive filtering techniques.

REFERENCES:
patent: 5052043 (1991-09-01), Gaborski
patent: 5129039 (1992-07-01), Hiraiwa
patent: 5140670 (1992-08-01), Chua et al.
patent: 5159660 (1992-10-01), Lu et al.
patent: 5168262 (1992-12-01), Okayama
patent: 5175678 (1992-12-01), Frerichs et al.
patent: 5253329 (1993-10-01), Villarreal et al.
patent: 5268834 (1993-12-01), Sanner et al.
patent: 5280564 (1994-01-01), Shiomi et al.
patent: 5282261 (1994-01-01), Skeirik
patent: 5371834 (1994-12-01), Tawel
patent: 5636326 (1997-06-01), Stork et al.
Chen H. and Liu R., An On-Line Unsupervised Learning Machine for Adaptive Feature Extraction, IEEE trans. on Circuits and Systems II, vol. 41, No. 2, pp. 87-98, (1994).
De Vries B., Gradient-Based Adaptation of Network Structure, International Conference on Artificial Neural Networks 94, Sorrento, Italy, (May 1994).
King S.Y., and Diamantaras K., A neural Network Learning algorithm for Adaptive Principal Component Extraction (APEX), Proc. of IEEE Int'l Conf. on Acoustics, Speech and Signal Processing (ICASSP), pp. 861-864, (1990).
Levin, A., and Leen T., and Moody J., Fast Pruning Using Principal Components, Advances in Neural Information Processing Systems (NIPS) 6, pp. 35-42, Morgan Kaufmann Publishers, (1994).
Levin et al., "Fast Pruning Using Principal Components," Advances in Neural Information Processing Systems (NIPS) 6, pp. 35-42, 1994.
Kung, S.Y. and Diamantaras, K., "A Neural Network Learning Algorithm for Adaptive Principal Component Extraction (APEX)", Proc. of IEEE Int'l Conf. on Acoustics, Speech and Signal Processing (ICASSP), pp. 861-864, 1990.
Levin, A., and Leen, T. and Moody, J., "Fast Pruning Using Principal Components", Advances in Neural Information Processing Systems (NIPS) 6, pp. 35-42, 1994.

Affiliated with

Also associated with

Rating

Method and system for training a neural network with adaptive we does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with Method and system for training a neural network with adaptive we, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method and system for training a neural network with adaptive we will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFUS-PAI-O-1635138