Ensembles of neural networks with different input sets

Details Ensembles of neural networks with different input sets Ensembles of neural networks with different input sets

2005-06-24

2009-11-03

Vincent, David R (Department: 2129)

Data processing: artificial intelligence

Neural network

Learning task

C706S020000

Reexamination Certificate

active

07613665

ABSTRACT:
Methods of creating and using robust neural network ensembles are disclosed. Some embodiments take the form of computer-based methods that comprise receiving a set of available inputs; receiving training data; training at least one neural network for each of at least two different subsets of the set of available inputs; and providing at least two trained neural networks having different subsets of the available inputs as components of a neural network ensemble configured to transform the available inputs into at least one output. The neural network ensemble may be applied as a log synthesis method that comprises: receiving a set of downhole logs; applying a first subset of downhole logs to a first neural network to obtain an estimated log; applying a second, different subset of the downhole logs to a second neural network to obtain an estimated log; and combining the estimated logs to obtain a synthetic log.

REFERENCES:
patent: 3802259 (1974-04-01), Echels
patent: 3946226 (1976-03-01), Smith, Jr.
patent: 3975157 (1976-08-01), Smith
patent: 4055763 (1977-10-01), Antkiw
patent: 4122240 (1978-10-01), Banas et al.
patent: 4122339 (1978-10-01), Smith, Jr. et al.
patent: 4122340 (1978-10-01), Smith, Jr. et al.
patent: 4239965 (1980-12-01), Oliver et al.
patent: 4293933 (1981-10-01), Park et al.
patent: 4297575 (1981-10-01), Smith, Jr. et al.
patent: 4430567 (1984-02-01), Oliver et al.
patent: 4605854 (1986-08-01), Smith, Jr.
patent: 4617825 (1986-10-01), Ruhovets
patent: 4645926 (1987-02-01), Randall
patent: 4646240 (1987-02-01), Serra et al.
patent: 4656354 (1987-04-01), Randall
patent: 4912655 (1990-03-01), Wood
patent: 4926488 (1990-05-01), Nadas et al.
patent: 5112126 (1992-05-01), Graebner
patent: 5251286 (1993-10-01), Wiener et al.
patent: 5374823 (1994-12-01), Odom
patent: 5444619 (1995-08-01), Hoskins et al.
patent: 5461698 (1995-10-01), Schwanke et al.
patent: 5465321 (1995-11-01), Smyth
patent: 5469404 (1995-11-01), Barber et al.
patent: 5475509 (1995-12-01), Okamoto
patent: 5517854 (1996-05-01), Plumb et al.
patent: 5525797 (1996-06-01), Moake
patent: 5608215 (1997-03-01), Evans
patent: 5828981 (1998-10-01), Callender et al.
patent: 5848379 (1998-12-01), Bishop
patent: 5862513 (1999-01-01), Mezzatesta et al.
patent: 5870690 (1999-02-01), Frenkel et al.
patent: 5875284 (1999-02-01), Watanabe et al.
patent: 5900627 (1999-05-01), Odom et al.
patent: 5940777 (1999-08-01), Keskes
patent: 6044327 (2000-03-01), Goldman
patent: 6092017 (2000-07-01), Ishida et al.
patent: 6140816 (2000-10-01), Herron
patent: 6150655 (2000-11-01), Odom
patent: 6192352 (2001-02-01), Alouani et al.
patent: 6207953 (2001-03-01), Wilson
patent: 6216134 (2001-04-01), Heckerman et al.
patent: 6272434 (2001-08-01), Wisler et al.
patent: 6295504 (2001-09-01), Ye et al.
patent: 6311591 (2001-11-01), Grossmann et al.
patent: 6317730 (2001-11-01), Neuneier et al.
patent: 6374185 (2002-04-01), Taner et al.
patent: 6381591 (2002-04-01), Hoffmann et al.
patent: 6411903 (2002-06-01), Bush
patent: 6424956 (2002-07-01), Werbos
patent: 6456990 (2002-09-01), Hoffmann et al.
patent: 6466893 (2002-10-01), Latwesen et al.
patent: 6760716 (2004-07-01), Ganesamoorthi et al.
patent: 6789620 (2004-09-01), Schultz et al.
patent: 2002/0147695 (2002-10-01), Khedkar et al.
patent: 2002/0152030 (2002-10-01), Schultz et al.
patent: 2002/0165911 (2002-11-01), Gabber et al.
patent: 2002/0170022 (2002-11-01), Shirai et al.
patent: 2002/0177954 (2002-11-01), Vail, III
patent: 2002/0178150 (2002-11-01), Hytopoulos et al.
patent: 2002/0183932 (2002-12-01), West et al.
patent: 2002/0187469 (2002-12-01), Kolodner et al.
patent: 2002/0188424 (2002-12-01), Grinstein et al.
patent: 2003/0115164 (2003-06-01), Jeng et al.
patent: 2004/0019427 (2004-01-01), San Martin
patent: 2004/0117121 (2004-06-01), Gray et al.
patent: 2004/0133531 (2004-07-01), Chen et al.
patent: 2004/0222019 (2004-11-01), Estes et al.
patent: 2004/0257240 (2004-12-01), Chen et al.
patent: 2005/0246297 (2005-11-01), Chen et al.
patent: 2007/0011114 (2007-01-01), Chen et al.
patent: 2007/0011115 (2007-01-01), Smith, Jr. et al.
patent: 04089998 (1992-03-01), None
patent: WO-9964896 (1999-12-01), None
Quirein, J. A., et al., “An Assessment of Neural Networks Applied to Pulsed Neutron Data for Predicting Open Hole Triple Combo Data”, SPWLA 44th Annual Logging Symposium, 2003.
Helle, B.H., et al., “Fluid Saturation From Sell Logs Using Committe Neural Networks”, Petroluem Geoscience, 2002, vol. 8, pp. 109-118.
Mullen, M., et al., “The Application of Neural Networks to Improve the Usability of Pulsed Neutron Logs for Evaluating Infill Well Locations in the Piceance Basin of Western Colorado and the San Juan Basin of Northwest New Mexico”, SPE Rocky Mountain Technology Conference, 2001.
Helle et al., Hans, “Fluid Saturation from Well Logs Using Committe Neural Networks”, 2002.
Quirein et al., John, “An Assessment of Neural Networks Applied to Pulsed Neutron Data for Predicting Open Hole Triple Combo Data”, 2003.
Chen, Dingding et al., U.S. Appl. No. 10/811,403, filed Mar. 26, 2004, entitled “Genetic Algorithm Based Selection Of Neural Network Ensemble For Processing Well Logging Data,” pgs.
Everson, R. M. et al., Full Elite-Sets for Multi-objective Optimisation, 2002, Eight pages, Fifth International Conference on Adaptive Computing in Design and Manufacture (ACDM 2002).
Gaspar-Cunha, A., RPSGAe—Reduced Pareto Set Genetic Algorithm: A Multiobjective Genetic Algorithm with Elitism, Nov. 2002, Six pages, Workshop on Multiple Objective Metaheuristics, Carre des Sciences, Paris, France.
Hansen, L.K. et al., “Neural Network Ensembles,” IEEE Transactions On Pattern Analysis and Machine Intelligence, vol. 12, No. 10, 1990, pp. 993-1001.
Helle, B.H. et al., “Fluid Saturation From Well Logs Using Committee Neural Networks,” Petroleum Geoscience, 2002, vol. 8, pp. 109-118.
Islam, M. M. et al., “A Constructive Algorithm For Training Cooperative Neural Network Ensembles,” IEEE Transactions On Neural Networks, 2003, vol. 14, No. 4, pp. 820-834.
Krough, A. et al., “Neural Network Ensembles, Cross Validation, And Active Learning,” Advances In Neural Information Processing Systems 7, Cambridge, MA., MIT Press, 1995, pp. 231-238.
Liu, Y. et al., “Ensemble Learing Via Negative Correlation,” Neural Networks, vol. 12, 1999, pp. 1399-1404.
Mullen, M. et al., “The Application Of Neural Networks To Improve The Usability Of Pulsed Neutron Logs For Evaluating Infill Well Locations In The Piceance Basin Of Western Colorado And The San Juan Basin Of Northwest New Mexico,” SPE Rocky Mountain Petroleum Technology Conference, 2001, Keystone, Colorado, 14 pgs.
Quirein, J. A., et al., “An Assestment Of Neural Networks Applied To Pulsed Neutron Data For Predicting Open Hole Triple Combo Data,” SPWLA 44th Annual Logging Symposium, 2003, 14 pgs.
Zhou, Z. et al., “Genetic Algorithm Based Selective Neural Network Ensemble,” Proceedings Of The 17th International Joint Conference On Artificial Intelligence, vol. 2, 2001, pp. 797-802.
Streeter, R.W. et al., “Cased Hole Exploration: Modem Pulsed Neutron Techniques For Locating By-Passed Hydrocarbons In Old Wells,” SPE 35162, SPE Permian Basin Oil & Gas Recovery Conference, Mar. 27-29, 1996, pp. 167-176, Midland, Texas.
Anonymous, “Log Interpretation Charts,” Dresser Atlas, Dresser Industries, Inc., USA, Jun. 1983, 2 pages.
Tittman, J. et al., “The Physical Foundations Of Formation Density Logging (Gamma Gamma),” Goephysics, vol. XXX, No. 2, Apr. 1965, pp. 284-293.
Odom, Richard C. et al., “A New 1.625″ Diameter Pulsed Neutron Capture And Inelastic/Capture Spectral Combination System Provides Answers In Complex Reservoirs,” SPWLA 35th Annual Logging Symposium, Jun. 1994, 19 pages.
Odom, Richard C., Program And Pertinent Slides From A Presentation by Richard C. Odom at SIAM Symposium On Inverse Problems:

Affiliated with

Also associated with

Rating

Ensembles of neural networks with different input sets does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with Ensembles of neural networks with different input sets, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Ensembles of neural networks with different input sets will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFUS-PAI-O-4107349