Branislav Bosansky

Transiting areas patrolled by a mobile adversary

We study the problem of a mobile agent trying to cross an area patrolled by a mobile adversary. The transiting agent aims to choose its route so as to minimize the probability of hostile encounter; the patroller agent, controlling one or more patrol units, aims at the opposite. We model the problem as a two-player zero-sum game (termed transit game) and search for an optimum route selection strategy as a mixed Nash equilibrium of the game. In contrast to existing game-theoretic models of this kind, we explicitly consider the limited endurance of patrols and the notion of bases to which the patrols need to repeatedly return. Noting the prohibitive size of the transit game, we employ two techniques for reducing the complexity of finding Nash equilibria - a compact network-flow-based representation of transit routes and iterative single- and double-oracle algorithms for incremental game matrix construction. We measure the computational time of all the methods on a range of transit game instances. In order to assess the practical relevance of the approach, we apply the transit game model and its solution to the real-world case of ship transit through areas affected by piracy. The results obtained using an agent-based simulation of maritime traffic show that the randomized game-theoretic transit routing strategy results in a lower number of pirate attacks than the currently employed method based on static transit corridors.

Case Studies of Network Defense with Attack Graph Games

The increasing complexity of securing modern computer networks makes decision support systems an important tool for administrators. A challenge many existing tools fail to address is that attackers react strategically to new security measures, adapting their behaviors in response. Game theory provides a methodology for making decisions that takes into account these reactions, rather than assuming static attackers. The authors present an overview of how game theory can be used to inform one type of security decision: how to optimally place honeypots in a network. They demonstrate this approach on a realistic case study and present initial validation results based on a study comparing their approach with human decision makers.

Agent-based process-critiquing decision support system

Formalization of the work practice using processes is a widely used method throughout various fields of industry. However, their application within the healthcare domain is rather unsuccessful due to the need for high agility, exceptions handling, and for working with complex medical knowledge that affects these processes. To overcome these problems the agent paradigm and multi-agent systems can be applied. In this paper we present a novel architecture of a multi-agent system that is able to work with general processes. As an exemplary application of the architecture we describe a critiquing decision support system for healthcare specialists based on formalized medical guidelines.

Integration of procedural knowledge in multi-agent systems in medicine

In this paper we present two developed multi-agent architectures. The first one models knowledge-based support of the eServices in the domain of home care, while the second one represents a general model that uses procedural knowledge in the form of organizational processes and formalized medical guidelines in a decision support system. We show that integration of both architectures is feasible and moreover, it can enhance functionality of the home care system.