A Game-Theoretic Approach to Design Secure and Resilient Distributed Support Vector Machines

Rui Zhang, Quanyan Zhu

Research output: Contribution to journalArticle

Abstract

Distributed support vector machines (DSVMs) have been developed to solve large-scale classification problems in networked systems with a large number of sensors and control units. However, the systems become more vulnerable, as detection and defense are increasingly difficult and expensive. This paper aims to develop secure and resilient DSVM algorithms under adversarial environments in which an attacker can manipulate the training data to achieve his objective. We establish a game-theoretic framework to capture the conflicting interests between an adversary and a set of distributed data processing units. The Nash equilibrium of the game allows predicting the outcome of learning algorithms in adversarial environments and enhancing the resilience of the machine learning through dynamic distributed learning algorithms. We prove that the convergence of the distributed algorithm is guaranteed without assumptions on the training data or network topologies. Numerical experiments are conducted to corroborate the results. We show that the network topology plays an important role in the security of DSVM. Networks with fewer nodes and higher average degrees are more secure. Moreover, a balanced network is found to be less vulnerable to attacks.

Original languageEnglish (US)
JournalIEEE Transactions on Neural Networks and Learning Systems
DOIs
StateAccepted/In press - Mar 6 2018

Fingerprint

Support vector machines
Parallel algorithms
Learning algorithms
Topology
Learning systems
Sensors
Experiments

Keywords

  • Adversarial machine learning
  • distributed support vector machines (DSVMs)
  • game theory
  • Games
  • Heuristic algorithms
  • Machine learning algorithms
  • networked systems
  • resilience
  • Security
  • security.
  • Support vector machines
  • Training
  • Training data

ASJC Scopus subject areas

  • Software
  • Computer Science Applications
  • Computer Networks and Communications
  • Artificial Intelligence

Cite this

@article{f44481c9ba19472bafbaf2a384b4a441,
title = "A Game-Theoretic Approach to Design Secure and Resilient Distributed Support Vector Machines",
abstract = "Distributed support vector machines (DSVMs) have been developed to solve large-scale classification problems in networked systems with a large number of sensors and control units. However, the systems become more vulnerable, as detection and defense are increasingly difficult and expensive. This paper aims to develop secure and resilient DSVM algorithms under adversarial environments in which an attacker can manipulate the training data to achieve his objective. We establish a game-theoretic framework to capture the conflicting interests between an adversary and a set of distributed data processing units. The Nash equilibrium of the game allows predicting the outcome of learning algorithms in adversarial environments and enhancing the resilience of the machine learning through dynamic distributed learning algorithms. We prove that the convergence of the distributed algorithm is guaranteed without assumptions on the training data or network topologies. Numerical experiments are conducted to corroborate the results. We show that the network topology plays an important role in the security of DSVM. Networks with fewer nodes and higher average degrees are more secure. Moreover, a balanced network is found to be less vulnerable to attacks.",
keywords = "Adversarial machine learning, distributed support vector machines (DSVMs), game theory, Games, Heuristic algorithms, Machine learning algorithms, networked systems, resilience, Security, security., Support vector machines, Training, Training data",
author = "Rui Zhang and Quanyan Zhu",
year = "2018",
month = "3",
day = "6",
doi = "10.1109/TNNLS.2018.2802721",
language = "English (US)",
journal = "IEEE Transactions on Neural Networks and Learning Systems",
issn = "2162-237X",
publisher = "IEEE Computational Intelligence Society",

}

TY - JOUR

T1 - A Game-Theoretic Approach to Design Secure and Resilient Distributed Support Vector Machines

AU - Zhang, Rui

AU - Zhu, Quanyan

PY - 2018/3/6

Y1 - 2018/3/6

N2 - Distributed support vector machines (DSVMs) have been developed to solve large-scale classification problems in networked systems with a large number of sensors and control units. However, the systems become more vulnerable, as detection and defense are increasingly difficult and expensive. This paper aims to develop secure and resilient DSVM algorithms under adversarial environments in which an attacker can manipulate the training data to achieve his objective. We establish a game-theoretic framework to capture the conflicting interests between an adversary and a set of distributed data processing units. The Nash equilibrium of the game allows predicting the outcome of learning algorithms in adversarial environments and enhancing the resilience of the machine learning through dynamic distributed learning algorithms. We prove that the convergence of the distributed algorithm is guaranteed without assumptions on the training data or network topologies. Numerical experiments are conducted to corroborate the results. We show that the network topology plays an important role in the security of DSVM. Networks with fewer nodes and higher average degrees are more secure. Moreover, a balanced network is found to be less vulnerable to attacks.

AB - Distributed support vector machines (DSVMs) have been developed to solve large-scale classification problems in networked systems with a large number of sensors and control units. However, the systems become more vulnerable, as detection and defense are increasingly difficult and expensive. This paper aims to develop secure and resilient DSVM algorithms under adversarial environments in which an attacker can manipulate the training data to achieve his objective. We establish a game-theoretic framework to capture the conflicting interests between an adversary and a set of distributed data processing units. The Nash equilibrium of the game allows predicting the outcome of learning algorithms in adversarial environments and enhancing the resilience of the machine learning through dynamic distributed learning algorithms. We prove that the convergence of the distributed algorithm is guaranteed without assumptions on the training data or network topologies. Numerical experiments are conducted to corroborate the results. We show that the network topology plays an important role in the security of DSVM. Networks with fewer nodes and higher average degrees are more secure. Moreover, a balanced network is found to be less vulnerable to attacks.

KW - Adversarial machine learning

KW - distributed support vector machines (DSVMs)

KW - game theory

KW - Games

KW - Heuristic algorithms

KW - Machine learning algorithms

KW - networked systems

KW - resilience

KW - Security

KW - security.

KW - Support vector machines

KW - Training

KW - Training data

UR - http://www.scopus.com/inward/record.url?scp=85043359654&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85043359654&partnerID=8YFLogxK

U2 - 10.1109/TNNLS.2018.2802721

DO - 10.1109/TNNLS.2018.2802721

M3 - Article

C2 - 29993612

AN - SCOPUS:85043359654

JO - IEEE Transactions on Neural Networks and Learning Systems

JF - IEEE Transactions on Neural Networks and Learning Systems

SN - 2162-237X

ER -