Apparatus and methods for machine learning hypotheses

Details Apparatus and methods for machine learning hypotheses Apparatus and methods for machine learning hypotheses

1997-07-29

1998-10-06

MacDonald, Allen R.

Data processing: artificial intelligence

Neural network

Learning method

706 26, G06F 1518

Patent

active

058192476

ABSTRACT:
Apparatus and methods for machine learning the hypotheses used in the classifier component of pattern classification devices such as OCRs, other image analysis systems, and and text retrieval systems. The apparatus and methods employ machine learning techniques for generating weak hypotheses from a set of examples of the patterns to be recognized and then evaluate the resulting hypothesis against example patterns. The results of the evaluation are used to increase the probability that the examples used to generate the next weak hypothesis are ones which the previous weak hypothesis did not correctly classify. The results of the evaluation are also used to give a weight to each weak hypothesis. A strong hypothesis is then made by combining the weak hypotheses according to their weights.

REFERENCES:
patent: 5479575 (1995-12-01), Yoda
H. Drucker, R. Schapire, P.Simard, "Boosting Performance in Neural Networks", International Journal of Pattern Recognition and Artificial Intelligence, vol. 7, No. 4 (1993) 705-719.
R. Schapire, "The Strength of Weak Learnability", Machine Learning, 5, 197-227 (1990).
Y. Freund, "Boosting a Weak Learning Algorithm by Majority", Proceedings of the Third Annual Workshop on Computational Learning Theory, Aug. 6-8, 1990.
H. Drucker, C. Cortes, L. D. Jackel, Y. LeCun, V. Vapnik, "Boosting and Other Machine Learing Algorithms", Machine Learning, Proceedings of the Eleventh International Conference, Jul. 10-13, 1994.
J. Jackson, "An Efficient Membership-Query Algorithm for Learning DNF with Respect to the Uniform Distribution", Proceedings of the 35th Annual Symposium on Foundations of Computer Science, Nov. 20-22, 1994.
Zhang, B-T., "Selecting a Critical Subset of Given Examples during Learning", ICANN '94 Proc. of Int'l. Conf. on Artificial Neural Networks, Sorrento, Italy, 26-29 May 1994, ISBN 3-540-19887-3, Springer-Verlag, Berlin, Germany, pp. 517-520, vol. 1.
EP-A-0 463 630 (Digital Equipment Corp.) 2 Jan. 1992 abstract.
EP-A-0 528 399 (Toyoda Machine Works Ltd) 24 Feb. 1993, p. 4, lines 14-57.
EP-A-0 591 921 (Matsushita Electric Ind Co. Ltd) 13 Apr. 1994, p. 6, line 13, p. 7, line 1.
Gathercole, C., et al., "Dynamic Training Subset Selection for Supervised Learning in Genetic Programming", Third Conf. on Parallel Problem Solving From Nature. Int'l Conf. on Evolutionary Computation, Jerusalem, Israel Oct. 1994, ISDN 3-540-58484-6, 1994, Springer-Verlag, Berlin, Germany, pp. 312-321 abstract.

Affiliated with

Also associated with

Rating

Apparatus and methods for machine learning hypotheses does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with Apparatus and methods for machine learning hypotheses, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Apparatus and methods for machine learning hypotheses will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFUS-PAI-O-93377