skip to main content
OSTI.GOV title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: Propagating Mixed Uncertainties in Cyber Attacker Payoffs: Exploration of Two-Phase Monte Carlo Sampling and Probability Bounds Analysis

Abstract

Securing cyber-systems on a continual basis against a multitude of adverse events is a challenging undertaking. Game-theoretic approaches, that model actions of strategic decision-makers, are increasingly being applied to address cybersecurity resource allocation challenges. Such game-based models account for multiple player actions and represent cyber attacker payoffs mostly as point utility estimates. Since a cyber-attacker’s payoff generation mechanism is largely unknown, appropriate representation and propagation of uncertainty is a critical task. In this paper we expand on prior work and focus on operationalizing the probabilistic uncertainty quantification framework, for a notional cyber system, through: 1) representation of uncertain attacker and system-related modeling variables as probability distributions and mathematical intervals, and 2) exploration of uncertainty propagation techniques including two-phase Monte Carlo sampling and probability bounds analysis.

Authors:
; ; ;
Publication Date:
Research Org.:
Pacific Northwest National Lab. (PNNL), Richland, WA (United States)
Sponsoring Org.:
USDOE
OSTI Identifier:
1334867
Report Number(s):
PNNL-SA-120091
DOE Contract Number:
AC05-76RL01830
Resource Type:
Conference
Resource Relation:
Conference: IEEE Symposium on Technologies for Homeland Security (HST 2016), May 10-11, 2016, Waltham, MA
Country of Publication:
United States
Language:
English
Subject:
game theory; cyber security

Citation Formats

Chatterjee, Samrat, Tipireddy, Ramakrishna, Oster, Matthew R., and Halappanavar, Mahantesh. Propagating Mixed Uncertainties in Cyber Attacker Payoffs: Exploration of Two-Phase Monte Carlo Sampling and Probability Bounds Analysis. United States: N. p., 2016. Web. doi:10.1109/THS.2016.7568967.
Chatterjee, Samrat, Tipireddy, Ramakrishna, Oster, Matthew R., & Halappanavar, Mahantesh. Propagating Mixed Uncertainties in Cyber Attacker Payoffs: Exploration of Two-Phase Monte Carlo Sampling and Probability Bounds Analysis. United States. doi:10.1109/THS.2016.7568967.
Chatterjee, Samrat, Tipireddy, Ramakrishna, Oster, Matthew R., and Halappanavar, Mahantesh. Fri . "Propagating Mixed Uncertainties in Cyber Attacker Payoffs: Exploration of Two-Phase Monte Carlo Sampling and Probability Bounds Analysis". United States. doi:10.1109/THS.2016.7568967.
@article{osti_1334867,
title = {Propagating Mixed Uncertainties in Cyber Attacker Payoffs: Exploration of Two-Phase Monte Carlo Sampling and Probability Bounds Analysis},
author = {Chatterjee, Samrat and Tipireddy, Ramakrishna and Oster, Matthew R. and Halappanavar, Mahantesh},
abstractNote = {Securing cyber-systems on a continual basis against a multitude of adverse events is a challenging undertaking. Game-theoretic approaches, that model actions of strategic decision-makers, are increasingly being applied to address cybersecurity resource allocation challenges. Such game-based models account for multiple player actions and represent cyber attacker payoffs mostly as point utility estimates. Since a cyber-attacker’s payoff generation mechanism is largely unknown, appropriate representation and propagation of uncertainty is a critical task. In this paper we expand on prior work and focus on operationalizing the probabilistic uncertainty quantification framework, for a notional cyber system, through: 1) representation of uncertain attacker and system-related modeling variables as probability distributions and mathematical intervals, and 2) exploration of uncertainty propagation techniques including two-phase Monte Carlo sampling and probability bounds analysis.},
doi = {10.1109/THS.2016.7568967},
journal = {},
number = ,
volume = ,
place = {United States},
year = {Fri Sep 16 00:00:00 EDT 2016},
month = {Fri Sep 16 00:00:00 EDT 2016}
}

Conference:
Other availability
Please see Document Availability for additional information on obtaining the full-text document. Library patrons may search WorldCat to identify libraries that hold this conference proceeding.

Save / Share:
  • Representation and propagation of uncertainty in cyber attacker payoffs is a key aspect of security games. Past research has primarily focused on representing the defender’s beliefs about attacker payoffs as point utility estimates. More recently, within the physical security domain, attacker payoff uncertainties have been represented as Uniform and Gaussian probability distributions, and intervals. Within cyber-settings, continuous probability distributions may still be appropriate for addressing statistical (aleatory) uncertainties where the defender may assume that the attacker’s payoffs differ over time. However, systematic (epistemic) uncertainties may exist, where the defender may not have sufficient knowledge or there is insufficient information aboutmore » the attacker’s payoff generation mechanism. Such epistemic uncertainties are more suitably represented as probability boxes with intervals. In this study, we explore the mathematical treatment of such mixed payoff uncertainties.« less
  • Quantification and propagation of uncertainties in cyber attacker payoffs is a key aspect within multiplayer, stochastic security games. These payoffs may represent penalties or rewards associated with player actions and are subject to various sources of uncertainty, including: (1) cyber-system state, (2) attacker type, (3) choice of player actions, and (4) cyber-system state transitions over time. Past research has primarily focused on representing defender beliefs about attacker payoffs as point utility estimates. More recently, within the physical security domain, attacker payoff uncertainties have been represented as Uniform and Gaussian probability distributions, and mathematical intervals. For cyber-systems, probability distributions may helpmore » address statistical (aleatory) uncertainties where the defender may assume inherent variability or randomness in the factors contributing to the attacker payoffs. However, systematic (epistemic) uncertainties may exist, where the defender may not have sufficient knowledge or there is insufficient information about the attacker’s payoff generation mechanism. Such epistemic uncertainties are more suitably represented as generalizations of probability boxes. This paper explores the mathematical treatment of such mixed payoff uncertainties. A conditional probabilistic reasoning approach is adopted to organize the dependencies between a cyber-system’s state, attacker type, player actions, and state transitions. This also enables the application of probabilistic theories to propagate various uncertainties in the attacker payoffs. An example implementation of this probabilistic framework and resulting attacker payoff distributions are discussed. A goal of this paper is also to highlight this uncertainty quantification problem space to the cyber security research community and encourage further advancements in this area.« less
  • Abstract not provided.
  • The concept of ``progressive Lattice Sampling`` as a basis for generating successive finite element response surfaces that are increasingly effective in matching actual response functions is investigated here. The goal is optimal response surface generation, which achieves an adequate representation of system behavior over the relevant parameter space of a problem with a minimum of computational and user effort. Such is important in global optimization and in estimation of system probabilistic response, which are both made much more viable by replacing large complex computer models of system behavior by fast running accurate approximations. This paper outlines the methodology for Finitemore » Element/Lattice Sampling (FE/LS) response surface generation and examines the effectiveness of progressively refined FE/LS response surfaces in decoupled Monte Carlo analysis of several model problems. The proposed method is in all cases more efficient (generally orders of magnitude more efficient) than direct Monte Carlo evaluation, with no appreciable loss of accuracy. Thus, when arriving at probabilities or distributions by Monte Carlo, it appears to be more efficient to expend computer model function evaluations on building a FE/LS response surface than to expend them in direct Monte Carlo sampling. Furthermore, the marginal efficiency of the FE/LS decoupled Monte Carlo approach increases as the size of the computer model increases, which is a very favorable property.« less
  • There are numerous uncertainties in the prediction of core parameters of innovative reactor designs, arising from approximations used in the solution of the transport equation, geometrical models, and in nuclear data processing and cross section library generation. This paper describes the problems encountered in the analysis of the Encapsulated Nuclear Heat Source (ENHS) core benchmark and the new cross section libraries developed to overcome these problems. The ENHS is a lead-bismuth or lead cooled novel reactor concept that is fuelled with metallic alloy of Pu, U and Zr, and is designed to operate for 20 effective full power years withoutmore » re-fuelling and with very small burn-up reactivity swing. The computational tools used include: MOCUP - a coupled MCNP-4C and ORIGEN2.1 using MCNP data libraries based on ENDF/B-VI evaluations; and KWO2-a coupled KENO-V.a and ORIGEN2.1 using ENDFB-V.2 based 238-group library. Two geometrical models of the ENHS core were used: a detailed 2D-rz model and approximate 2D-xy mode. Uncertainties in the cross sections of lead were found to be particularly large and deserve careful evaluation. (authors)« less