On solving large-scale low-rank zero-sum security games of incomplete information

Amnol Monga, Quanyan Zhu

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Zero-sum games are useful to model the strategic interactions between an attacker and a defender over a network. The computation of the equilibrium solutions become challenging when entries of the payoff matrix are either partially or completely unknown. The uncertainties of the game often arise from the error in the observations and the incomplete information on the structure of the game. In this work, we leverage low-rank features of the security games to establish a robust computational framework that enables to solve games of incomplete information and compute their security strategies. We characterize the upper and the lower security values of the games using two robust and convex optimization problems in which the players compute the worst-case strategies subject to their unknown. The robust framework is compared with the benchmark matrix completion approach in which saddle-point equilibrium strategies are computed after the matrix is completed using rank optimization. Our numerical results have shown that the strategy found by using the holistic robust optimization method can outperform the matrix completion method in the prediction of the attacker and defender strategies.

Original languageEnglish (US)
Title of host publication8th IEEE International Workshop on Information Forensics and Security, WIFS 2016
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9781509011384
DOIs
StatePublished - Jan 18 2017
Event8th IEEE International Workshop on Information Forensics and Security, WIFS 2016 - Abu Dhabi, United Arab Emirates
Duration: Dec 4 2016Dec 7 2016

Other

Other8th IEEE International Workshop on Information Forensics and Security, WIFS 2016
CountryUnited Arab Emirates
CityAbu Dhabi
Period12/4/1612/7/16

Fingerprint

Convex optimization
Games of incomplete information
uncertainty
interaction
Values
Uncertainty
Equilibrium solution
Leverage
Incomplete information
Zero-sum game
Optimization problem
Prediction
Saddlepoint
Strategic interaction
Benchmark
Robust optimization

ASJC Scopus subject areas

  • Computer Networks and Communications
  • Information Systems and Management
  • Safety, Risk, Reliability and Quality
  • Law

Cite this

Monga, A., & Zhu, Q. (2017). On solving large-scale low-rank zero-sum security games of incomplete information. In 8th IEEE International Workshop on Information Forensics and Security, WIFS 2016 [7823923] Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/WIFS.2016.7823923

On solving large-scale low-rank zero-sum security games of incomplete information. / Monga, Amnol; Zhu, Quanyan.

8th IEEE International Workshop on Information Forensics and Security, WIFS 2016. Institute of Electrical and Electronics Engineers Inc., 2017. 7823923.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Monga, A & Zhu, Q 2017, On solving large-scale low-rank zero-sum security games of incomplete information. in 8th IEEE International Workshop on Information Forensics and Security, WIFS 2016., 7823923, Institute of Electrical and Electronics Engineers Inc., 8th IEEE International Workshop on Information Forensics and Security, WIFS 2016, Abu Dhabi, United Arab Emirates, 12/4/16. https://doi.org/10.1109/WIFS.2016.7823923
Monga A, Zhu Q. On solving large-scale low-rank zero-sum security games of incomplete information. In 8th IEEE International Workshop on Information Forensics and Security, WIFS 2016. Institute of Electrical and Electronics Engineers Inc. 2017. 7823923 https://doi.org/10.1109/WIFS.2016.7823923
Monga, Amnol ; Zhu, Quanyan. / On solving large-scale low-rank zero-sum security games of incomplete information. 8th IEEE International Workshop on Information Forensics and Security, WIFS 2016. Institute of Electrical and Electronics Engineers Inc., 2017.
@inproceedings{f86c6c72bee44c03a01dee25c24dc52e,
title = "On solving large-scale low-rank zero-sum security games of incomplete information",
abstract = "Zero-sum games are useful to model the strategic interactions between an attacker and a defender over a network. The computation of the equilibrium solutions become challenging when entries of the payoff matrix are either partially or completely unknown. The uncertainties of the game often arise from the error in the observations and the incomplete information on the structure of the game. In this work, we leverage low-rank features of the security games to establish a robust computational framework that enables to solve games of incomplete information and compute their security strategies. We characterize the upper and the lower security values of the games using two robust and convex optimization problems in which the players compute the worst-case strategies subject to their unknown. The robust framework is compared with the benchmark matrix completion approach in which saddle-point equilibrium strategies are computed after the matrix is completed using rank optimization. Our numerical results have shown that the strategy found by using the holistic robust optimization method can outperform the matrix completion method in the prediction of the attacker and defender strategies.",
author = "Amnol Monga and Quanyan Zhu",
year = "2017",
month = "1",
day = "18",
doi = "10.1109/WIFS.2016.7823923",
language = "English (US)",
booktitle = "8th IEEE International Workshop on Information Forensics and Security, WIFS 2016",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
address = "United States",

}

TY - GEN

T1 - On solving large-scale low-rank zero-sum security games of incomplete information

AU - Monga, Amnol

AU - Zhu, Quanyan

PY - 2017/1/18

Y1 - 2017/1/18

N2 - Zero-sum games are useful to model the strategic interactions between an attacker and a defender over a network. The computation of the equilibrium solutions become challenging when entries of the payoff matrix are either partially or completely unknown. The uncertainties of the game often arise from the error in the observations and the incomplete information on the structure of the game. In this work, we leverage low-rank features of the security games to establish a robust computational framework that enables to solve games of incomplete information and compute their security strategies. We characterize the upper and the lower security values of the games using two robust and convex optimization problems in which the players compute the worst-case strategies subject to their unknown. The robust framework is compared with the benchmark matrix completion approach in which saddle-point equilibrium strategies are computed after the matrix is completed using rank optimization. Our numerical results have shown that the strategy found by using the holistic robust optimization method can outperform the matrix completion method in the prediction of the attacker and defender strategies.

AB - Zero-sum games are useful to model the strategic interactions between an attacker and a defender over a network. The computation of the equilibrium solutions become challenging when entries of the payoff matrix are either partially or completely unknown. The uncertainties of the game often arise from the error in the observations and the incomplete information on the structure of the game. In this work, we leverage low-rank features of the security games to establish a robust computational framework that enables to solve games of incomplete information and compute their security strategies. We characterize the upper and the lower security values of the games using two robust and convex optimization problems in which the players compute the worst-case strategies subject to their unknown. The robust framework is compared with the benchmark matrix completion approach in which saddle-point equilibrium strategies are computed after the matrix is completed using rank optimization. Our numerical results have shown that the strategy found by using the holistic robust optimization method can outperform the matrix completion method in the prediction of the attacker and defender strategies.

UR - http://www.scopus.com/inward/record.url?scp=85015027496&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85015027496&partnerID=8YFLogxK

U2 - 10.1109/WIFS.2016.7823923

DO - 10.1109/WIFS.2016.7823923

M3 - Conference contribution

BT - 8th IEEE International Workshop on Information Forensics and Security, WIFS 2016

PB - Institute of Electrical and Electronics Engineers Inc.

ER -