The importance of research and applications related to security and multi-agent systems continues to increase in a broad variety of disciplines, including computer science, electrical engineering, economics, biology, political science, business, law, public policy, and many others. The focus of this workshop is to bring together the broad community working on Security and Multi-Agent Systems motivated by any of these domains.

Many large-scale real-world security problems have been successfully modeled as multi-agent security games, and highly scalable algorithmic solutions with software assistants implementing these have been developed and deployed. Perhaps the greatest advances have been in the domain of physical security, with examples including patrolling of seaports and airports, scheduling air marshals, ticket audit in transit systems. Remarkably, there have been a number of more recent developments that significantly broaden the applicability of security game approaches. For example, similar techniques have been used in important sustainability applications, such as fishery protection and prevention of illegal poaching. Moreover, game theoretic models have increasing applicability in cyber, as well as cyber-physical system (CPS) security, such as adversarial machine learning methods (for use, for example, in intrusion detection systems), resilient sensor placement and monitoring strategies, and privacy preserving data publishing and auditing systems.

While there has been significant progress, there still exist many major challenges facing the design of effective approaches to deal with the difficulties in real-world domains. These include building predictive behavioral models for the players, dealing with uncertainties in games, scaling up for large games, and applications of machine learning and multi-agent learning to security, particularly in the context of repeated or stochastic games. This workshop is structured to encourage a lively exchange of ideas to address the above challenges.

We invite full length research submissions from a broad range of researchers and practitioners, including (1) computer scientists applying their AI/MAS research to real-world security problems, (2) interdisciplinary researchers combining AI/MAS with various disciplines (e.g., game theory, operations research, social science, and psychology), and (3) engineers and scientists from private companies and public organizations performing security related research and development, as well as building real adversarial reasoning systems. We encourage all researchers working towards applying security and multi-agent systems concepts for real-world problems to submit to the workshop.

Topics of interest include, but are not limited to:

Real-world applications of game theory for security
Cybersecurity
Security applications of machine learning
Foundations of game theory for security
Adversarial/robust learning
Online learning
Learning in games
Algorithms for scaling to very large games
Economics of security
Behavioral game theory
Decision making under uncertainty
Agent/human interaction for preference elicitation and optimization
Protection against environmental crime
Risk analysis and modeling
Security applications of AI methods
Evaluation/lessons learned of deployed systems

Submissions should be anonymized. Submissions are due by Feb 1, 2016 (midnight, Pacific time). The page length and format are as follows:

Full-length papers (up to 8 pages in AAMAS format, excluding references)