Adversarial Classification When Even Good Types Can Fake

6 Pages Posted: 14 Jan 2009

See all articles by Asunur Cezar

Asunur Cezar

affiliation not provided to SSRN

Srinivasan Raghunathan

University of Texas at Dallas - Naveen Jindal School of Management

Sumit Sarkar

University of Texas at Dallas - Department of Information Systems & Operations Management

Date Written: December 8, 2007

Abstract

In some classification domains, firms face agents who actively manipulate their information to mislead the firm about their true types so as to avoid unfavorable decisions as a result of the classification. In such domains, firms should take the possibility of applicants' faking behavior into consideration in their decision making. We consider situations where the firm faces agents who can modify instances regardless of their type; unlike prior work, we don't restrict ourselves to those situations where only malicious agents manipulate their data. We show that the firm is never better off when agents have the ability to fake than when they do not. However, surprisingly, a reduction in faking cost does not always hurt the firm, implying that a firm may sometimes prefer an environment in which agents can fake more easily over another in which it is more difficult to fake.

Suggested Citation

Cezar, Asunur and Raghunathan, Srinivasan and Sarkar, Sumit, Adversarial Classification When Even Good Types Can Fake (December 8, 2007). Available at SSRN: https://ssrn.com/abstract=1327852 or http://dx.doi.org/10.2139/ssrn.1327852

Asunur Cezar

affiliation not provided to SSRN ( email )

Srinivasan Raghunathan

University of Texas at Dallas - Naveen Jindal School of Management ( email )

P.O. Box 830688
Richardson, TX 75083-0688
United States

Sumit Sarkar (Contact Author)

University of Texas at Dallas - Department of Information Systems & Operations Management ( email )

P.O. Box 830688
Richardson, TX 75083-0688
United States
972-883-6854 (Phone)
972-883-6811 (Fax)

Do you have a job opening that you would like to promote on SSRN?

Paper statistics

Downloads
58
Abstract Views
849
rank
418,553
PlumX Metrics