Risk Management in game situations: Principal-agent and adversarial examples
نویسنده
چکیده
This paper presents and illustrates several cases of risk analysis involving games in two different kinds of situations. The first is a management case in which the principal sets the rules of the game for an agent managing the development of a subsystem and if late may choose to take shortcuts. The second involves a class of adversarial situations, illustrated here by the analysis of counter-insurgency policies, and of US nuclear counter-proliferation strategies in the face of a country’s weapons development efforts such as those of North Korea and Iran. These situations involve two different kinds of games and result in various risks to the main decision maker. The models presented here are designed to support these decisions and are based on systems analysis, Bayesian probability and game analysis and/or simulation.
منابع مشابه
ارائه مدلی جهت استفاده ازعاملهای متحرک در سیستم های تشخیص نفوذ توزیع شده مبتنی بر تئوری بازی
The proposed framework applies two game theoretic models for economic deployment of intrusion detection system (IDS). The first scheme models and analyzes the interaction behaviors of between an attacker and intrusion detection agent within a non-cooperative game, and then the security risk value is derived from the mixed strategy Nash equilibrium. The second scheme uses the security risk value...
متن کاملA Generative Game-Theoretic Framework for Adversarial Plan Recognition
Adversarial reasoning is of the first importance for defence and security applications since it allows to (1) better anticipate future threats, and (2) be proactive in deploying effective responses. In this paper, we address the two subtasks of adversarial reasoning, namely adversarial plan recognition and strategy formulation, from a generative, game-theoretic perspective. First, a set of poss...
متن کاملRisk management in urban tunnels using methods of game theory and multi-criteria decision-making
In general, underground spaces are associated with high risks because of their high uncertainty in geotechnical environments. Since most accidents and incidents in these structures are often associated with uncertainty, the development of risk analysis and management methods and prevention of accidents are essential. A deeper recognition of the factors affecting the implementation process can p...
متن کاملAdversarial classification: An adversarial risk analysis approach
Classification problems in security settings are usually contemplated as confrontations in which one or more adversaries try to fool a classifier to obtain a benefit. Most approaches to such adversarial classification problems have focused on game theoretical ideas with strong underlying common knowledge assumptions, which are actually not realistic in security domains. We provide an alternativ...
متن کاملDetecting Adversarial Attacks on Neural Network Policies with Visual Foresight
Deep reinforcement learning has shown promising results in learning control policies for complex sequential decision-making tasks. However, these neural network-based policies are known to be vulnerable to adversarial examples. This vulnerability poses a potentially serious threat to safety-critical systems such as autonomous vehicles. In this paper, we propose a defense mechanism to defend rei...
متن کامل