A PARTIALLY LINEAR FRAMEWORK FOR MASSIVE HETEROGENEOUS DATA
A PARTIALLY LINEAR FRAMEWORK FOR MASSIVE HETEROGENEOUS DATA
About this item
Full title
Author / Creator
Zhao, Tianqi , Cheng, Guang and Liu, Han
Publisher
United States: Institute of Mathematical Statistics
Journal title
Language
English
Formats
Publication information
Publisher
United States: Institute of Mathematical Statistics
Subjects
More information
Scope and Contents
Contents
We consider a partially linear framework for modeling massive heterogeneous data. The major goal is to extract common features across all subpopulations while exploring heterogeneity of each subpopulation. In particular, we propose an aggregation type estimator for the commonality parameter that possesses the (nonasymptotic) minimax optimal bound a...
Alternative Titles
Full title
A PARTIALLY LINEAR FRAMEWORK FOR MASSIVE HETEROGENEOUS DATA
Authors, Artists and Contributors
Author / Creator
Identifiers
Primary Identifiers
Record Identifier
TN_cdi_pubmedcentral_primary_oai_pubmedcentral_nih_gov_5394596
Permalink
https://devfeature-collection.sl.nsw.gov.au/record/TN_cdi_pubmedcentral_primary_oai_pubmedcentral_nih_gov_5394596
Other Identifiers
ISSN
0090-5364
E-ISSN
2168-8966
DOI
10.1214/15-AOS1410
