Abstract
Distributed machine learning algorithms play a significant role in processing massive data sets over large networks. However, the increasing reliance on machine learning on information and communication technologies makes it inherently vulnerable to cyber threats. This work aims to develop secure distributed algorithms to protect the learning from adversaries. We establish a game-theoretic framework to capture the conflicting goals of a learner who uses distributed support vector machines (DSVM) and an attacker who is capable of flipping training labels. We develop a fully distributed and iterative algorithm to capture real-time reactions of the learner at each node to adversarial behaviors. The numerical results show that DSVM is vulnerable to attacks, and their impact has a strong dependence on the network topologies.
Original language | English (US) |
---|---|
Title of host publication | 2017 51st Annual Conference on Information Sciences and Systems, CISS 2017 |
Publisher | Institute of Electrical and Electronics Engineers Inc. |
ISBN (Electronic) | 9781509047802 |
DOIs | |
State | Published - May 10 2017 |
Event | 51st Annual Conference on Information Sciences and Systems, CISS 2017 - Baltimore, United States Duration: Mar 22 2017 → Mar 24 2017 |
Other
Other | 51st Annual Conference on Information Sciences and Systems, CISS 2017 |
---|---|
Country | United States |
City | Baltimore |
Period | 3/22/17 → 3/24/17 |
Fingerprint
ASJC Scopus subject areas
- Signal Processing
- Information Systems and Management
- Computer Networks and Communications
- Information Systems
Cite this
A game-theoretic analysis of label flipping attacks on distributed support vector machines. / Zhang, Rui; Zhu, Quanyan.
2017 51st Annual Conference on Information Sciences and Systems, CISS 2017. Institute of Electrical and Electronics Engineers Inc., 2017. 7926118.Research output: Chapter in Book/Report/Conference proceeding › Conference contribution
}
TY - GEN
T1 - A game-theoretic analysis of label flipping attacks on distributed support vector machines
AU - Zhang, Rui
AU - Zhu, Quanyan
PY - 2017/5/10
Y1 - 2017/5/10
N2 - Distributed machine learning algorithms play a significant role in processing massive data sets over large networks. However, the increasing reliance on machine learning on information and communication technologies makes it inherently vulnerable to cyber threats. This work aims to develop secure distributed algorithms to protect the learning from adversaries. We establish a game-theoretic framework to capture the conflicting goals of a learner who uses distributed support vector machines (DSVM) and an attacker who is capable of flipping training labels. We develop a fully distributed and iterative algorithm to capture real-time reactions of the learner at each node to adversarial behaviors. The numerical results show that DSVM is vulnerable to attacks, and their impact has a strong dependence on the network topologies.
AB - Distributed machine learning algorithms play a significant role in processing massive data sets over large networks. However, the increasing reliance on machine learning on information and communication technologies makes it inherently vulnerable to cyber threats. This work aims to develop secure distributed algorithms to protect the learning from adversaries. We establish a game-theoretic framework to capture the conflicting goals of a learner who uses distributed support vector machines (DSVM) and an attacker who is capable of flipping training labels. We develop a fully distributed and iterative algorithm to capture real-time reactions of the learner at each node to adversarial behaviors. The numerical results show that DSVM is vulnerable to attacks, and their impact has a strong dependence on the network topologies.
UR - http://www.scopus.com/inward/record.url?scp=85020223548&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85020223548&partnerID=8YFLogxK
U2 - 10.1109/CISS.2017.7926118
DO - 10.1109/CISS.2017.7926118
M3 - Conference contribution
AN - SCOPUS:85020223548
BT - 2017 51st Annual Conference on Information Sciences and Systems, CISS 2017
PB - Institute of Electrical and Electronics Engineers Inc.
ER -