Patent
1996-08-23
1998-03-31
Hafiz, Tariq R.
395 21, 395 24, G06E 100, G06F 1518
Patent
active
057347972
ABSTRACT:
A system for generating a minimal artificial neural network (ANN) architecture having as inputs the minimal number of features necessary to discriminate between event classes. A network generator generates an initial ANN architecture. A training processor generates a trained ANN by training the initial ANN to a desired degree of accuracy. A pruning processor prunes the trained ANN by removing interconnections and nodes from the trained ANN. The training processor and pruning processor continue to train and prune the ANN until a minimal network architecture having the class discrimination features as its only inputs is obtained.
REFERENCES:
patent: 5444824 (1995-08-01), Austvold et al.
patent: 5546503 (1996-08-01), Abe et al.
patent: 5636326 (1997-06-01), Stork et al.
Jain et al. "Learning Texture Discrimination Masks," IEEE Transactions on ttern Analysis and Machine Intelligence, vol. 18, Issue 2, pp. 195-205, Feb. 29, 1996.
Lursinap "Applied Self-Recovery Technique to Link and Neuron Prunings," Circuits and Systems, 1994 Midwest Symposium, V. 1, pp. 545-548, Dec. 31, 1994.
Luk et al. "Pruning via Number of Links and Neuron Activities," Circuits and Systems, 1993 IEEE International Symposium, pp. 2415-2418, May 31, 1993.
Khorasani et al. "Structure Adaptation in FeedFoward Neural Networks," Neural Networks, 1994 International Conf., vol. 3, pp. 1403-1408, Dec. 31, 1994.
Giles et al. "Pruning Recurrent Neural Networks for Improved Generalization Performance," IEEE Transaction on Neural Networks, vol. 5, Issue 5, pp. 848-851, Sep. 30, 1994.
Deangelis Christopher M.
Green Robert W.
Eipert William F.
Hafiz Tariq R.
Lall Prithvi C.
McGowan Michael J.
The United States of America as represented by the Secretary of
LandOfFree
System and method for determining class discrimination features does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with System and method for determining class discrimination features, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and System and method for determining class discrimination features will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-59052