Sorry, you need to enable JavaScript to visit this website.

facebooktwittermailshare

Communication efficient coreset sampling for distributed learning

Abstract: 

In this paper, distributed learning is studied using the approach of coreset. In the context of classification, an algorithm of coreset construction is proposed to reduce the redundancy of data and thus the communication requirement, similarly to source coding in traditional data communications. It is shown that the coreset based boosting has a high convergence rate and small sample complexity. Moreover, it is robust to adversary distribution, thus leading to potential applications in distributed learning systems. Both theoretical and numerical analyses are provided to demonstrate the proposed framework.

up
0 users have voted:

Paper Details

Authors:
Yawen Fan, Husheng Li
Submitted On:
20 June 2018 - 9:57am
Short Link:
Type:
Poster
Event:
Presenter's Name:
Yawen Fan
Document Year:
2018
Cite

Document Files

poster_v.pdf

(53 downloads)

Subscribe

[1] Yawen Fan, Husheng Li, "Communication efficient coreset sampling for distributed learning", IEEE SigPort, 2018. [Online]. Available: http://sigport.org/3258. Accessed: Nov. 12, 2018.
@article{3258-18,
url = {http://sigport.org/3258},
author = {Yawen Fan; Husheng Li },
publisher = {IEEE SigPort},
title = {Communication efficient coreset sampling for distributed learning},
year = {2018} }
TY - EJOUR
T1 - Communication efficient coreset sampling for distributed learning
AU - Yawen Fan; Husheng Li
PY - 2018
PB - IEEE SigPort
UR - http://sigport.org/3258
ER -
Yawen Fan, Husheng Li. (2018). Communication efficient coreset sampling for distributed learning. IEEE SigPort. http://sigport.org/3258
Yawen Fan, Husheng Li, 2018. Communication efficient coreset sampling for distributed learning. Available at: http://sigport.org/3258.
Yawen Fan, Husheng Li. (2018). "Communication efficient coreset sampling for distributed learning." Web.
1. Yawen Fan, Husheng Li. Communication efficient coreset sampling for distributed learning [Internet]. IEEE SigPort; 2018. Available from : http://sigport.org/3258