Show simple item record

dc.contributor.authorDing, Lizhong
dc.contributor.authorLiu, Zhi
dc.contributor.authorLi, Yu
dc.contributor.authorLiao, Shizhong
dc.contributor.authorLiu, Yong
dc.contributor.authorYang, Peng
dc.contributor.authorYu, Ge
dc.contributor.authorShao, Ling
dc.contributor.authorGao, Xin
dc.date.accessioned2019-01-15T13:34:22Z
dc.date.available2019-01-15T13:34:22Z
dc.date.issued2019-09-07
dc.identifier.citationDing, L., Liu, Z., Li, Y., Liao, S., Liu, Y., Yang, P., … Gao, X. (2019). Linear Kernel Tests via Empirical Likelihood for High-Dimensional Data. Proceedings of the AAAI Conference on Artificial Intelligence, 33, 3454–3461. doi:10.1609/aaai.v33i01.33013454
dc.identifier.doi10.1609/aaai.v33i01.33013454
dc.identifier.urihttp://hdl.handle.net/10754/630864
dc.description.abstractWe propose a framework for analyzing and comparing distributions without imposing any parametric assumptions via empirical likelihood methods. Our framework is used to study two fundamental statistical test problems: the two-sample test and the goodness-of-fit test. For the two-sample test, we need to determine whether two groups of samples are from different distributions; for the goodness-of-fit test, we examine how likely it is that a set of samples is generated from a known target distribution. Specifically, we propose empirical likelihood ratio (ELR) statistics for the two-sample test and the goodness-of-fit test, both of which are of linear time complexity and show higher power (i.e., the probability of correctly rejecting the null hypothesis) than the existing linear statistics for high-dimensional data. We prove the nonparametric Wilks’ theorems for the ELR statistics, which illustrate that the limiting distributions of the proposed ELR statistics are chi-square distributions. With these limiting distributions, we can avoid bootstraps or simulations to determine the threshold for rejecting the null hypothesis, which makes the ELR statistics more efficient than the recently proposed linear statistic, finite set Stein discrepancy (FSSD). We also prove the consistency of the ELR statistics, which guarantees that the test power goes to 1 as the number of samples goes to infinity. In addition, we experimentally demonstrate and theoretically analyze that FSSD has poor performance or even fails to test for high-dimensional data. Finally, we conduct a series of experiments to evaluate the performance of our ELR statistics as compared to state-of-the-art linear statistics.
dc.description.sponsorshipThis publication is based upon work supported by the King Abdullah University of Science and Technology (KAUST) Office of Sponsored Research (OSR) under Award No. URF/1/3007-01-01 and BAS/1/1624-01-01, National Natural Science Foundation of China (No. 61673293), National Natural Science Foundation of China (No. 61703396) and Shenzhen Government (GJHZ20180419190732022).
dc.publisherAssociation for the Advancement of Artificial Intelligence (AAAI)
dc.relation.urlhttps://aaai.org/ojs/index.php/AAAI/article/view/4222
dc.rightsArchived with thanks to Proceedings of the AAAI Conference on Artificial Intelligence
dc.titleLinear Kernel Tests via Empirical Likelihood for High-Dimensional Data
dc.typeArticle
dc.contributor.departmentComputer Science Program
dc.contributor.departmentComputational Bioscience Research Center (CBRC)
dc.contributor.departmentComputer, Electrical and Mathematical Sciences and Engineering (CEMSE) Division
dc.identifier.journalProceedings of the AAAI Conference on Artificial Intelligence
dc.conference.dateJanuary 27, 2019 – February 1, 2019
dc.conference.nameThe Thirty-Third AAAI Conference on Artificial Intelligence (AAAI-19)
dc.conference.locationHonolulu, U.S.
dc.eprint.versionPost-print
dc.contributor.institutionInception Institute of Artificial Intelligence (IIAI), Abu Dhabi, UAE
dc.contributor.institutionUniversity of Macau, China,
dc.contributor.institutionTianjin University, China
dc.contributor.institutionInstitute of Information Engineering, CAS, China
dc.contributor.institutionTechnology and Engineering Center for Space Utilization, CAS, China
kaust.personLi, Yu
kaust.personYang, Peng
kaust.personGao, Xin
kaust.grant.numberBAS/1/1624-01-01
kaust.grant.numberURF/1/3007-01-01
refterms.dateFOA2019-11-14T14:18:09Z
kaust.acknowledged.supportUnitOffice of Sponsored Research (OSR)


Files in this item

Thumbnail
Name:
lkt-0831.pdf
Size:
530.0Kb
Format:
PDF
Description:
Accepted Manuscript

This item appears in the following Collection(s)

Show simple item record