Analysis of Kernel Mean Matching under Covariate Shift

作者: Yao-liang Yu , Csaba Szepesv ri

DOI:

关键词:

摘要: In real supervised learning scenarios, it is not uncommon that the training and test sample follow different probability distributions, thus rendering necessity to correct sampling bias. Focusing on a particular covariate shift problem, we derive high confidence bounds for kernel mean matching (KMM) estimator, whose convergence rate turns out depend some regularity measure of regression function also capacity kernel. By comparing KMM with natural plug-in establish superiority former hence provide concrete evidence/ understanding effectiveness under shift.

参考文章(23)
Felipe Cucker, Ding Xuan Zhou, Learning Theory: An Approximation Theory Viewpoint ,(2007)
Ingo Steinwart, On the influence of the kernel on the consistency of support vector machines Journal of Machine Learning Research. ,vol. 2, pp. 67- 93 ,(2002) , 10.1162/153244302760185252
Corinna Cortes, Mehryar Mohri, Michael Riley, Afshin Rostamizadeh, Sample Selection Bias Correction Theory Lecture Notes in Computer Science. pp. 38- 53 ,(2008) , 10.1007/978-3-540-87987-9_8
Steve Smale, Ding-Xuan Zhou, None, Learning Theory Estimates via Integral Operators and Their Approximations Constructive Approximation. ,vol. 26, pp. 153- 172 ,(2007) , 10.1007/S00365-006-0659-Y
Takafumi Kanamori, Taiji Suzuki, Masashi Sugiyama, Statistical analysis of kernel-based least-squares density-ratio estimation Machine Learning. ,vol. 86, pp. 335- 367 ,(2012) , 10.1007/S10994-011-5266-3
N. Aronszajn, Theory of Reproducing Kernels. Transactions of the American Mathematical Society. ,vol. 68, pp. 337- 404 ,(1950) , 10.1090/S0002-9947-1950-0051437-7
Iosif Pinelis, OPTIMUM BOUNDS FOR THE DISTRIBUTIONS OF MARTINGALES IN BANACH SPACES Annals of Probability. ,vol. 22, pp. 1679- 1706 ,(1994) , 10.1214/AOP/1176988477
Bianca Zadrozny, Learning and evaluating classifiers under sample selection bias Twenty-first international conference on Machine learning - ICML '04. pp. 114- ,(2004) , 10.1145/1015330.1015425
Hidetoshi Shimodaira, Improving predictive inference under covariate shift by weighting the log-likelihood function Journal of Statistical Planning and Inference. ,vol. 90, pp. 227- 244 ,(2000) , 10.1016/S0378-3758(00)00115-4