Machine Learning Cyberattack and Defense Strategies

Bland, John A.; Petty, Mikel D.; Whitaker, Tymaine S.; Maxwell, Katia P.; Cantrell, Walter Alan

doi:10.1016/j.cose.2020.101738

Cited by 50 publications

(38 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Collectively, this leads to the need to simulate data when dealing with cybersecurity applications. This is explicitly stated in [20] corroborated by [8], [15]- [17], [17], [19], [22], [23]. These abstractions have the objective of portraying several complexities from a real cyber environment and can provide game like environments where RL techniques can be deployed to evaluate defence strategies.…”

Section: Agentmentioning

confidence: 84%

Prospective Artificial Intelligence Approaches for Active Cyber Defence

Dhir,

Hoeltgebaum,

Adams

et al. 2021

Preprint

View full text Add to dashboard Cite

Cyber criminals are rapidly developing new malicious tools that leverage artificial intelligence (AI) to enable new classes of adaptive and stealthy attacks. New defensive methods need to be developed to counter these threats. Some cyber security professionals are speculating AI will enable corresponding new classes of active cyber defence measures -is this realistic, or currently mostly hype? The Alan Turing Institute, with expert guidance from the UK National Cyber Security Centre and Defence Science Technology Laboratory, published a research roadmap for AI for ACD last year. This position paper updates the roadmap for two of the most promising AI approachesreinforcement learning and causal inference -and describes why they could help tip the balance back towards defenders.

show abstract

Section: Agentmentioning

confidence: 84%

Prospective Artificial Intelligence Approaches for Active Cyber Defence

Dhir,

Hoeltgebaum,

Adams

et al. 2021

Preprint

View full text Add to dashboard Cite

show abstract

“…In contrast, RL does not require large data sets; instead, in RL the data is generated during the learning process in the form of rewards for actions taken, which serve as evaluative feedback to a learning agent [16,17]. RL has increased in importance in recent years because of its applicability in a wide range of disciplines for which identifying all possible examples to train the agent is difficult; those applications include robotics, gaming, computer vision, business, control system engineering, simulation-based optimization, networks, anti-crime operations [18] and cybersecurity [19].…”

Section: Reinforcement Learningmentioning

confidence: 99%

“…The weights and costs of the reward response variables were also the same as for factory layout 1 (see Table 4). Figures 18,19,20,21,22 and 23 show the results of the experiments for the six initial plans; there is one figure for each of the initial plans. As before, in each figure, the x axis represents the number of plan change actions taken and the y axis represents the MH cost of a plan after each action.…”

Section: (Line 15) a Lower Action Valuementioning

confidence: 99%

Applying reinforcement learning to plan manufacturing material handling

Govindaiah¹,

Petty

2021

Discov Artif Intell

Self Cite

View full text Add to dashboard Cite

Applying machine learning methods to improve the efficiency of complex manufacturing processes, such as material handling, can be challenging. The interconnectedness of the multiple components that make up real-world manufacturing processes and the typically very large number of variables required to specify procedures and plans within them combine to make it very difficult to map the details of such processes to a formal mathematical representation suitable for conventional optimization methods. Instead, in this work reinforcement learning was applied to produce increasingly efficient plans for material handling in representative manufacturing facilities. Doing so included defining a formal representation of a realistically complex material handling plan, specifying a set of suitable plan change operators as reinforcement learning actions, implementing a simulation-based multi-objective reward function that considers multiple components of material handling costs, and abstracting the many possible material handling plans into a state set small enough to enable reinforcement learning. Experimentation with multiple material handling plans on two separate factory layouts indicated that reinforcement learning could consistently reduce the cost of material handling. This work demonstrates the applicability of reinforcement learning with a multi-objective reward function to realistically complex material handling processes.

show abstract

“…This is due to several reasons. First, it is problematic to identify breaches in companies (Bland et al 2020;Wang and Zhang 2020;Seibold et al 2020;Sahoo and Gupta 2019;Heartfield et al 2018;Choo 2011). Most of the existing research has focused on investigating a certain typology of attacks and on studying cases of companies, without considering a wider spectrum of possible types of attacks that companies may receive (Garre, Pérez, and Ruiz-Martínez 2021;Dahiya and Gupta 2020).…”

Section: Introductionmentioning

confidence: 99%

The severity and effects of Cyber-breaches in SMEs: a machine learning approach

Arroyabe

Arróyabe

2021

Enterprise Information Systems

View full text Add to dashboard Cite

In this paper, we investigate cyber breaches and their effects on small and medium enterprises (SMEs), considering the role that cybersecurity plays in SMEs, and the importance that SMEs have in the economy. Using the Cyber Security Breaches Survey data, the first contribution extends previous works confirming that SMEs receive a wide variety of breaches. Secondly, we have characterized the degree of severity of breaches in SMEs, based on disruption time and their cost. Our last contribution consists of determining the effect and severity of breaches in SMEs in terms of economic, financial and management impacts.

show abstract

Machine Learning Cyberattack and Defense Strategies

Cited by 50 publications

References 7 publications

Prospective Artificial Intelligence Approaches for Active Cyber Defence

Prospective Artificial Intelligence Approaches for Active Cyber Defence

Applying reinforcement learning to plan manufacturing material handling

The severity and effects of Cyber-breaches in SMEs: a machine learning approach

Contact Info

Product

Resources

About