No Access

REGULARIZED LEAST SQUARE ALGORITHM WITH TWO KERNELS

School of Mathematic Science, University of Jinan, Shandong Provincial Key Laboratory of Network-Based Intelligent Computing, Jinan 250022, P. R. China

Search for more papers by this author

and

PING LIU

No. 1 Middle School of LiJing, Dongying 257400, P. R. China

Search for more papers by this author

https://doi.org/10.1142/S0219691312500439Cited by:2 (Source: Crossref)

Abstract

A new multi-kernel regression learning algorithm is studied in this paper. In our setting, the hypothesis space is generated by two Mercer kernels, thus it has stronger approximation ability than the single kernel case. We provide the mathematical foundation for this regularized learning algorithm. We obtain satisfying capacity-dependent error bounds and learning rates by the covering number method.

Keywords:

AMSC: 68T05, 68Q32, 62J02

References

N. Aronszajn, Trans. Amer. Math. Soc. 68, 337 (1950), DOI: 10.1090/S0002-9947-1950-0051437-7. Crossref, Web of Science, Google Scholar
O. Bousquet and A. Elisseeff, J. Mach. Learning Research 2, 499 (2002). Web of Science, Google Scholar
B. Bakker and T. Heskes, J. Mach. Learning Research 4, 83 (2003). Web of Science, Google Scholar
F. Bauer, S. Pereverzev and L. Rosasco, J. Complexity 23, 52 (2007), DOI: 10.1016/j.jco.2006.07.001. Crossref, Web of Science, Google Scholar
R. Caruana, Machine Learning 28, 41 (1997), DOI: 10.1023/A:1007379606734. Crossref, Web of Science, Google Scholar
X. Y. Changet al., Int. J. Wavelets, Multiresolut. Inf. Process. 9(4), 549 (2011), DOI: 10.1142/S0219691311004213. Link, Web of Science, Google Scholar
F. Cucker and S. Smale, Bull. Amer. Math. Soc. 39, 1 (2001), DOI: 10.1090/S0273-0979-01-00923-5. Crossref, Web of Science, Google Scholar
F. Cucker and S. Smale, Found. Comput. Math. 2(4), 413 (2002), DOI: 10.1007/s102080010030. Crossref, Web of Science, Google Scholar
O. Chapelleet al., Machine Learning 46, 131 (2002), DOI: 10.1023/A:1012450327387. Crossref, Web of Science, Google Scholar
R. G. Douglas , Banach Algebra Techniques in Operator Theory ( Springer , 1998 ) . Crossref, Google Scholar
T. Evgeniou and M. Pontil, Regularized multi-task learning, Proceedings of the Tenth ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (2004) pp. 109–117. Google Scholar
T. Evgeniou, M. Pontil and T. Poggio, Adv. Comput. Math. 13, 1 (2000), DOI: 10.1023/A:1018946025316. Crossref, Web of Science, Google Scholar
J. Li et al. , Mixture Density Estimation , Advances in Neural Information Processing Systems 12 ( The MIT Press , 2000 ) . Google Scholar
G. R. G. Lanckrietet al., J. Mach. Learning Research 5, 27 (2004). Web of Science, Google Scholar
T. Poggio and S. Smale, Notices Amer. Math. Soc. 50, 537 (2003). Google Scholar
A. Rakhlin, D. Panchenko and S. Mukherjee, ESAIM: Probability and Statistics 9, 222 (2005), DOI: 10.1051/ps:2005011. Google Scholar
S. Smale and D. X. Zhou, Constr. Approx. 26, 153 (2007), DOI: 10.1007/s00365-006-0659-y. Crossref, Web of Science, Google Scholar
S. Smale and D. X. Zhou, Appl. Comput. Harm. Anal. 19, 285 (2005), DOI: 10.1016/j.acha.2005.03.001. Crossref, Web of Science, Google Scholar
H. W. Sun and Q. Wu, Math. Comput. Model. 49, 276 (2009), DOI: 10.1016/j.mcm.2008.08.005. Crossref, Web of Science, Google Scholar
Q. Wu, Y. Ying and D.-X. Zhou, Found Comput. Math. 6, 171 (2006), DOI: 10.1007/s10208-004-0155-9. Crossref, Web of Science, Google Scholar