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

facebooktwittermailshare

Robust Distributed Gradient Descent with Arbitrary Number of Byzantine Attackers

Abstract: 

Due to the grow of modern dataset size and the desire to harness computing power of multiple machines, there is a recent surge of interest in the design of distributed machine learning algorithms. However, distributed algorithms are sensitive to Byzantine attackers who can send falsified data to prevent the convergence of algorithms or lead the algorithms to converge to value of the attackers' choice. Our novel algorithm can deal with an arbitrary number of Byzantine attackers.

up
0 users have voted:

Paper Details

Authors:
Lifeng Lai
Submitted On:
14 April 2018 - 12:40am
Short Link:
Type:
Poster
Event:
Presenter's Name:
Xinyang Cao
Paper Code:
1822
Document Year:
2018
Cite

Document Files

ICASSPposter.pdf

(339 downloads)

Subscribe

[1] Lifeng Lai, "Robust Distributed Gradient Descent with Arbitrary Number of Byzantine Attackers", IEEE SigPort, 2018. [Online]. Available: http://sigport.org/2795. Accessed: Oct. 17, 2018.
@article{2795-18,
url = {http://sigport.org/2795},
author = {Lifeng Lai },
publisher = {IEEE SigPort},
title = {Robust Distributed Gradient Descent with Arbitrary Number of Byzantine Attackers},
year = {2018} }
TY - EJOUR
T1 - Robust Distributed Gradient Descent with Arbitrary Number of Byzantine Attackers
AU - Lifeng Lai
PY - 2018
PB - IEEE SigPort
UR - http://sigport.org/2795
ER -
Lifeng Lai. (2018). Robust Distributed Gradient Descent with Arbitrary Number of Byzantine Attackers. IEEE SigPort. http://sigport.org/2795
Lifeng Lai, 2018. Robust Distributed Gradient Descent with Arbitrary Number of Byzantine Attackers. Available at: http://sigport.org/2795.
Lifeng Lai. (2018). "Robust Distributed Gradient Descent with Arbitrary Number of Byzantine Attackers." Web.
1. Lifeng Lai. Robust Distributed Gradient Descent with Arbitrary Number of Byzantine Attackers [Internet]. IEEE SigPort; 2018. Available from : http://sigport.org/2795