A Sensitivity Analysis of Evolutionary Algorithms in Generating Secure Configurations

Dass, Shuvalaxmi; Namin, Akbar Siami

doi:10.1109/bigdata50022.2020.9378307

Cited by 4 publications

(3 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The closest research work to address the automatic generation of secure configurations is related to the application of evolutionary algorithms to this problem [9,10,17,18]. Such greedy algorithms are used to maximize the optimality of solutions.…”

Section: Discussionmentioning

confidence: 99%

“…According to our experience with evolutionary algorithms [17,18], techniques such as Genetic Algorithms (GA) and Particle Swarm Optimizations (PSO) are very effective when the search space is small; whereas, Reinforcement Learning is more viable for large search spaces due to the fact they need large amount of training data to be effective. As a result, the evolutionary algorithms may result in generating less diverse set of populations and samples and thus some of the samples might be repeated.…”

Section: Discussionmentioning

confidence: 99%

“…However, the choice of technique employed for a problem explicitly depends on the nature of the problem being solved. Evolutionary algorithms and techniques such as Genetic Algorithms (GA) and Particle Swarm Optimization (PSO) [17,18] have already been applied to secure configuration problems. In addition to these evolutionary algorithms, it is possible to adapt machine learning-based techniques to the problem and understand the pros and cons of these approaches.…”

Section: Why Reinforcement Learning For Generating Secure Configuration?mentioning

confidence: 99%

See 2 more Smart Citations

Reinforcement Learning for Generating Secure Configurations

Dass¹,

Namin²

2021

Electronics

Self Cite

View full text Add to dashboard Cite

Many security problems in software systems are because of vulnerabilities caused by improper configurations. A poorly configured software system leads to a multitude of vulnerabilities that can be exploited by adversaries. The problem becomes even more serious when the architecture of the underlying system is static and the misconfiguration remains for a longer period of time, enabling adversaries to thoroughly inspect the software system under attack during the reconnaissance stage. Employing diversification techniques such as Moving Target Defense (MTD) can minimize the risk of exposing vulnerabilities. MTD is an evolving defense technique through which the attack surface of the underlying system is continuously changing. However, the effectiveness of such dynamically changing platform depends not only on the goodness of the next configuration setting with respect to minimization of attack surfaces but also the diversity of set of configurations generated. To address the problem of generating a diverse and large set of secure software and system configurations, this paper introduces an approach based on Reinforcement Learning (RL) through which an agent is trained to generate the desirable set of configurations. The paper reports the performance of the RL-based secure and diverse configurations through some case studies.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Discussionmentioning

confidence: 99%

Section: Why Reinforcement Learning For Generating Secure Configuration?mentioning

confidence: 99%

See 1 more Smart Citation