Method and apparatus for training a neural network

Details Method and apparatus for training a neural network Method and apparatus for training a neural network

1995-01-11

1997-09-02

Davis, George B.

395 24, G06F 1518

Patent

active

056640676

ABSTRACT:
A plurality of training inputs are selected, wherein each training input corresponds to a first possible output. The quality of each of the plurality of training inputs is characterized. A first training input is selected from the training inputs, where the first training input is of higher quality than a second training input of the training inputs. The neural network is trained with the higher-quality first training input prior to training with the second training input. A neuron may be added to the neural network in accordance with the first training input, wherein the neuron is associated with the first possible output.

REFERENCES:
patent: 3701095 (1972-10-01), Yamaguchi et al.
patent: 3950733 (1976-04-01), Cooper et al.
patent: 4044243 (1977-08-01), Cooper et al.
patent: 4326259 (1982-04-01), Cooper et al.
patent: 4599693 (1986-07-01), Denenberg
patent: 4954963 (1990-09-01), Penz et al.
patent: 5010512 (1991-04-01), Hartstein et al.
patent: 5033006 (1991-07-01), Ishizuka et al.
patent: 5046020 (1991-09-01), Filkin
patent: 5054093 (1991-10-01), Cooper et al.
patent: 5058180 (1991-10-01), Khan
patent: 5060278 (1991-10-01), Fukumizu
patent: 5067164 (1991-11-01), Denker et al.
patent: 5086479 (1992-02-01), Takenaga et al.
patent: 5119438 (1992-06-01), Ueda et al.
patent: 5140530 (1992-08-01), Guha et al.
patent: 5150323 (1992-09-01), Castelaz
patent: 5175796 (1992-12-01), Refregier
patent: 5181171 (1993-01-01), McCormack et al.
patent: 5214715 (1993-05-01), Carpenter et al.
patent: 5214744 (1993-05-01), Schweizer et al.
patent: 5218646 (1993-06-01), Sirat et al.
patent: 5220618 (1993-06-01), Sirat et al.
patent: 5239594 (1993-08-01), Yoda
patent: 5245697 (1993-09-01), Suzuoka
patent: 5247584 (1993-09-01), Krogmann
patent: 5251268 (1993-10-01), Colley et al.
patent: 5260871 (1993-11-01), Goldberg
patent: 5265192 (1993-11-01), McCormack
patent: 5293454 (1994-03-01), Kamiya
patent: 5293456 (1994-03-01), Guez et al.
patent: 5297237 (1994-03-01), Masuoka et al.
patent: 5299284 (1994-03-01), Roy
patent: 5311601 (1994-05-01), Carpenter et al.
patent: 5313559 (1994-05-01), Ogata et al.
patent: 5335291 (1994-08-01), Kramer
patent: 5337371 (1994-08-01), Sato et al.
patent: 5339818 (1994-08-01), Baker et al.
patent: 5426745 (1995-06-01), Baji et al.
patent: 5479575 (1995-12-01), Yoda
patent: 5493688 (1996-02-01), Weingard
patent: 5500905 (1996-03-01), Martin et al.
Minnix et al, "A Multi-Layered Self-Organizing Artificial Neural Network for Invariant Pattern Recognition", IEEE Transactions on Knowledge & Data Engineering, vol. 4, No. 2, Apr. 1992 pp. 162-167.
Khotanzad et al, "Distortion Invariant Character Recognition by a Multi-Layer Perception and Back-Popagation Learning", IEEE Int'l Conf on Neural Networks, Jul. 24-27 1988, pp. 625-632 vol. 1.
Wen et al, "Self-Generating Neural Networks", Int'l Joint Conf on Neural Networks, Jun. 7-11 1992, pp. 850-855 vol. 4.
Lee et al, "Implementing a Self-development Neural Network Using Doubly Linked List", Proc 13th Ann. Int'l Computer SW and Applications Conf, Sep. 20-22 1989, pp. 672-679.
S. Gruber, "Neural network based inspection of machined surfaces using laser scattering," Industrial Inspection II, vol. 1265, 12 Mar. 1990, The Hague, NL, pp. 85-94.
1990 IEEE Transactions on Systems, Man, and Cybernetics, vol. 20, No. 4, Jul./Aug. 1990, pp. 816-825, authors Yuzo Hirai and Yasuyuki Tsukui, Title "Position Independent Pattern Matching by Neural Network".
1991 IEEE 10th Annual Int'l. Phoenix Conf. on Computers & Comm., Mar. 27-30 1991, pp. 39-45, authors Dae Su Kim and Terrance L. Huntsberger, Title "Self-organizing Neural Networks for Unsupervised Pattern Recognition".
1991 IEEE Int. Symp. on Circuits and Systems, vol. 1/5, 11 Jun. 1991, Singapore, pp. 356-359, XP000384785, authors Yoshikazu Miyanaga, Makoto Teraoka, and Koji Tochinai, Title "Parallel and Adaptive Clustering Method Suitable for a VLSI System".
Traitement Du Signal, vol. 8, No. 6, 1991, Paris, FR, pp. 423-430, XP000360600, authors Joel Minot and Philippe Gentric, Title "Authentification dynamique de signatures par reseaux de neurones".
1991 IEEE Int. Symp. on Circuits and Systems, vol. 2/5, 11 Jun. 1991, Singapore, pp. 1176-1179, XP000370414, authors I-Chang Jou et al, Title "A Hyperellipsoid Neural Network for Pattern Classification Section".
Pattern Recognition Letters, vol. 13, No. 5, May 1992, Amsterdam, NL, pp. 325-329, XP000278617, authors Gek Sok Lim et al; Title "Adaptive quadratic neural nets".
Omatu et al., "Neural Network Model for Alphabetical Letter Recognition," International Neural Network Conference, pp. 19-22, Paris, Jul. 9-13, 1990.
Specht, "Probabalistic Neural Networks for Classification, Mapping, or Associative Memory," 1988 IEEE International Conference on Neural Networks, pp. I-525-532.
Namatame, "A Connectionist Learning with High-Order Functional Networks and Its Internal Representation," Tools for Artificial Intelligence, pp. 542-547, Oct. 1989.
Reilly et al., "Learning System Architectures Composed of Multiple Learning Modules," IEEE First Int'l. Conf. on Neural Networks, pp. II-495-503, Jun. 1987.
Rumelhart et al., "Learning Internal Representation by Error Propagation," Parallel Distributed Processing, vol. I, pp. 318-362, 1986.
"Learning Systems Based on Multiple Neural Networks," Nestor, Inc. Providence, R.I.
Bachmann et al., "A relaxation model for memory with high storage density," Proc. Natl. Acad. Sci. USA, vol. 84, pp. 7529-7531, Nov. 1987, Biophysics.
Reilly et al., "A Neural Model for Category Learning," Biological Cybernetics, pp. 35-41, 1982.
Scofield et al., "Pattern Class Degeneracy in an Unrestricted Storage Density Memory," 1987 IEEE Conference on Neural Information Processing Systems, Nov. 1987.
Kavuri, "Solving the Hidden Node Problem in Networks with Ellipsoidal Units and Related Issues," pp. I-775-780, Jun. 8, 1992.
Kelly et al., "An Adaptive Algorithm for Modifying Hyperellipsoidal Decision Surfaces," pp. IV-196-201, Jun. 11, 1992.

Affiliated with

Also associated with

Rating

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

Rate now

Comments { 0 }

Profile ID: LFUS-PAI-O-314736