Robust personalization through biased regularization

Data processing: software development – installation – and managem – Software program development tool – Modeling

Reexamination Certificate

Rate now

  [ 0.00 ] – not rated yet Voters 0   Comments 0

Details

Reexamination Certificate

active

07886266

ABSTRACT:
The subject disclosure pertains to systems and methods for personalization of a recognizer. In general, recognizers can be used to classify input data. During personalization, a recognizer is provided with samples specific to a user, entity or format to improve performance for the specific user, entity or format. Biased regularization can be utilized during personalization to maintain recognizer performance for non-user specific input. In one aspect, regularization can be biased to the original parameters of the recognizer, such that the recognizer is not modified excessively during personalization.

REFERENCES:
patent: 5644652 (1997-07-01), Bellegarda et al.
patent: 5680480 (1997-10-01), Beernink et al.
patent: 5991742 (1999-11-01), Tran
patent: 6304667 (2001-10-01), Reitano
patent: 6725199 (2004-04-01), Brittan et al.
patent: 6842767 (2005-01-01), Partovi et al.
patent: 6842877 (2005-01-01), Robarts et al.
patent: 6886008 (2005-04-01), Blayvas et al.
patent: 6909708 (2005-06-01), Krishnaswamy et al.
patent: 6963832 (2005-11-01), Vanhilst
patent: 7239731 (2007-07-01), Semenov et al.
patent: 7469062 (2008-12-01), Napper
patent: 2004/0218232 (2004-11-01), Fishman
patent: 2005/0089227 (2005-04-01), Haluptzok et al.
patent: 2005/0099398 (2005-05-01), Garside et al.
patent: 2007/0140561 (2007-06-01), Abdulkader et al.
patent: 2009/0263043 (2009-10-01), Cristobal Perez et al.
Bishop, C.; Nueral Networks for Pattern Recognition; Ocford; Clarendon Press; New York; Oxford University Press 1995; 495 pages.
B. Scholkopf, et al., A Generalized Representer Theorem, D. Helmbold and B. Williamson (Eds.): COLT/ EuroCOLT 2001, LNAI 2111, pp. 416-426, 2001. Springer-Verlag Berlin Heidelberg 2001; 10 pages.
W. Sarle, Neural Network FAQ, What is Weight Decay?, Topic: Weight decay for neural networks, May 17, 2002, Available at http://www.faqs.org/faqs/ai-faq
eural-nets/part3/section-6.html, last accessed Oct. 14, 2006, 2 pages.
C. Chang, et al., LIBSVM: A Library for Support Vector Machines, 2001. Available at http://www.csie.ntu.edu.tw/˜cjin/libsvm; 22 pages.
C. Hsu, et al., A Comparison on Methods for Multi-Class Support Vector Machines, IEEE Transactions on Neural Networks, Mar. 2002, vol. 13, No. 2, pp. 415-425; 11 pages.
W. Kienzle, et al., Training Support Vector Machines with Multiple Equality Constraints, in 16th European Conference on Machine Learning, J. Gama et al. (Eds.): ECML 2005, LNAI 3720, Springer-Verlag Berlin Heidelberg 2005, pp. 182-193; 12 pages.
Y. Lecun, et al., Gradient-Based Learning Applied to Document Recognition, Proceedings of the IEEE, Nov. 1998, vol. 86, No. 11, pp. 2278-2324; 47 pages.
H.A. Rowley, et al., The Effect of Large Training Set Sizes on Online Japanese Kanji and English Cursive Recognizers, Proceedings of the 8th International Workshop on Frontiers in Handwriting Recognition (IWFHR), 2002, 5 pages; 5 pages.
P.Y. Simard, et al., Best Practices for Convolutional Neural Networks Applied to Visual Document Analysis, Advances in Neural Information Processing Systems 12, 2000, 6 pages.
P. Sollich, Probabilistic Methods for Support Vector Machines, in Advances in Neural Information Processing Systems 12, S.A. Solla et al. (Eds.), 2000, MIT Press, pp. 349-355; 7 pages.

LandOfFree

Say what you really think

Search LandOfFree.com for the USA inventors and patents. Rate them and share your experience with other people.

Rating

Robust personalization through biased regularization does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with Robust personalization through biased regularization, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Robust personalization through biased regularization will most certainly appreciate the feedback.

Rate now

     

Profile ID: LFUS-PAI-O-2629505

  Search
All data on this website is collected from public sources. Our data reflects the most accurate information available at the time of publication.