NetSE: Small: Beyond Tit-for-Tat: New Techniques for Collaboration in Network Security Games

University Of New Mexico, Albuquerque NM

Investigators

Abstract

Beyond Tit-for-Tat: New Techniques for Collaboration in Network Security Games Motivation and Problem: How can we ensure collaboration on the Internet, where populations are highly fluctuating, selfish, and unpredictable? This project will explore a new algorithmic technique for enabling collaboration in network security games. The new technique, Secure Multiparty Mediation (SMM), improves on past approaches such as tit-for-tat in the following ways: (1) it works even in single round games; (2) it works even when the actions of the players of the game are never revealed; (3) it works even in the presence of churn, i.e. players joining and leaving the game. In the SMM approach, advice is generated by using a mediator: an algorithm that generates private advice for each player. After receiving advice, each player decides on an action. The players retain free-will and so will follow the advice only if it is their best interest to do so. Preliminary work shows that this approach significantly improves social welfare in a well-studied network security game, the virus inoculation game. The SMM approach does not require the exchange of money or other resources. Moreover, the mediator can be implemented in a completely distributed fashion using algorithmic techniques from distributed computing. Intellectual Merit: This project will push the frontier of knowledge about collaboration in game theory. The concept of a mediator has been known in the economics community for many years. However, the project will result in mediators that are more sophisticated algorithmically than those proposed in the economics literature: they will be specifically designed to work for large-scale networks, and will use tools from distributed computing, cryptography and randomized algorithms. Moreover, the project will use analytical tools from the algorithmic game theory community to quantify the success of the mediators designed. Broader Impact: Success in this project will lead to development of a new mathematical tool for addressing security problems. The power of collaborative techniques will likely increase as the world's information infrastructure becomes increasingly connected. Thus, such a tool could significantly improve the security of the world's computer networks. Research results from this proposal will be incorporated into lectures and a project in a security class that has just been created in the researcher's department. Game theory is beguiling to students and attracts attention in popular culture (cf. the movie ?A Beautiful Mind"). Thus, success in incorporating game theory into a security class will likely attract and retain more talented students into security research. Game theory is also an important conduit between academic disciplines. Significant mathematical results in this area are likely to be of interest to other disciplines such as economics and biology. New Mexico is an EPSCoR state and the University of New Mexico is one of only two universities in the nation that is both a Minority Serving Institution and a Carnegie Very High Research Activity university. The researcher will make every effort to include women and minority students in this project.