arrow
Return

ADAPTIVE DISTRIBUTED METHODS UNDER COMMUNICATION CONSTRAINTS

delete2020-08-01
delete10
delete
OA
AI
B
Botond Szabó *
H
Harry van Zanten
DOI:10.1214/19-AOS1890delete
deleteOriginal
deleteOriginal request for help
deleteShare
deleteSave
Abstract

Abstract

En 中文
We study estimation methods under communication constraints in a distributed version of the nonparametric random design regression model. We derive minimax lower bounds and exhibit methods that attain those bounds. Moreover, we show that adaptive estimation is possible in this setting.
Keywords:
Distributed computation
minimax rates
adaptation
nonparametric regression
communication constraints

Journal

Annals of Statistics cover
Annals of Statistics
IF:
3.7
Papers:
2.8K
Citations:
2.9W

Organization

L
leiden university - excl lumc
Scholars:
3.5W
Papers: 2.9W
Citations: 46
L
Leiden University
Scholars:
4.0W
Papers: 3.3W
Citations: 3.8W
Cited Papers

Cited Papers

Genome restructuring in mouse embryos during reprogramming and early development
err2006-04-01
err0
errOAAI
errCatherine Martin; Nathalie Beaujean; Vincent Brochard; Christophe Audouard; Daniele Zink; Pascale Debey
errShare
errSave
A scalable bootstrap for massive data
err2014-03-17
err286
errOAAI
errKleiner, Ariel; Talwalkar, Ameet; Sarkar, Purnamrita; Jordan, Michael I.
errShare
errSave
Wetting phenomena for liquid Cu and Cu alloys on Fe-base substrate
err2016-05-01
err0
PREAI
errE. Bernardo; R. Oro; M. Campos; J.M. Torralba
errShare
errSave
Surgical management of cervical ganglioneuromas in children
err2006-02-01
err0
PREAI
errSteven B. Cannady; Brian J. Chung; Keiko Hirose; Noel Garabedian; Thierry Van Den Abbeele; Peter J. Koltai
errShare
errSave
Molecular detection and characterization of Shigella spp. harboring extended-spectrum β-lactamase genes in children with diarrhea in northwest Iran
err2022-12-08
err0
errOAAI
errSahar Sabour; Amir Teimourpour; Jafar Mohammadshahi; Hadi Peeridogaheh; Roghayeh Teimourpour; Taher Azimi; Zahra Hosseinali
errShare
errSave
researcher View more