A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2019; you can also visit the original URL. The file type is application/pdf
.
Filters
Adversarial Reinforcement Learning for Observer Design in Autonomous Systems under Cyber Attacks [article]
2018 arXiv pre-print
We develop a framework for using adversarial deep reinforcement learning to design observer strategies that are robust to adversarial errors in information channels. ... Complex autonomous control systems are subjected to sensor failures, cyber-attacks, sensor noise, communication channel failures, etc. that introduce errors in the measurements. ... Motivated by disparate strands of research on deep reinforcement learning and cyber attacks on autonomous systems, we investigate in this paper if deep reinforcement learning can be used to reliably design ...
arXiv:1809.06784v1 fatcat:q2xse7fhdjbmbaie4zcqwqorza
Adversarial Reinforcement Learning under Partial Observability in Autonomous Computer Network Defence [article]
2020 arXiv pre-print
While most existing work studies the problem in the context of computer vision or console games, this paper focuses on reinforcement learning in autonomous cyber defence under partial observability. ... Recent studies have demonstrated that reinforcement learning (RL) agents are susceptible to adversarial manipulation, similar to vulnerabilities previously demonstrated in the supervised learning setting ... We next abstract the threat model for adversarial learning in autonomous cyber defence as follows: Black-box approach. ...
arXiv:1902.09062v3 fatcat:4qwowgr7a5hsripc2nzvrmps4m
CybORG: An Autonomous Cyber Operations Research Gym [article]
2020 arXiv pre-print
Autonomous Cyber Operations (ACO) involves the consideration of blue team (defender) and red team (attacker) decision-making models in adversarial scenarios. ... Driven by the need to efficiently support reinforcement learning to train adversarial decision-making models through simulation and emulation, our design differs from prior related work. ... Design This section presents the overall design for CybORG (see Figure 1 ) incorporating simulation and emulation with a common API to support reinforcement learning for autonomous and adversarial cyber ...
arXiv:2002.10667v2 fatcat:b75jsmeobbg7taxjna3rhnb4ou
Informing Autonomous Deception Systems with Cyber Expert Performance Data [article]
2021 arXiv pre-print
This paper discusses methods for improving the realism and ecological validity of AI used for autonomous cyber defense by exploring the potential to use Inverse Reinforcement Learning (IRL) to gain insight ... defense system. ... Autonomous cyber defense also requires automating decision making, which is well-suited for artificial intelligence (AI) solutions-reinforcement learning (RL) in particular, can address many of the challenges ...
arXiv:2109.00066v1 fatcat:to6ihgt6mvcmdo4tf6hv5qi6yq
Autonomous Cyber Defense Introduces Risk: Can We Manage the Risk? [article]
2022 arXiv pre-print
Here we focus on machine learning training, algorithmic feedback, and algorithmic constraints, with the aim of motivating a discussion on achieving trust in autonomous cyber defenses. ... Autonomous agents have the potential to use ML with large amounts of data about known cyberattacks as input, in order to learn patterns and predict characteristics of future attacks. ... Autonomous Cyberdefense Introduces Risk: Can We Manage the Risk?. Computer, 54 (10) , 106-110. …………………………. ...
arXiv:2201.11148v1 fatcat:77kbpj7tcna7ndsnveotw4m6nm
Deep Reinforcement Learning for Cyber System Defense under Dynamic Adversarial Uncertainties [article]
2023 arXiv pre-print
Our results suggest the efficacy of DRL algorithms for proactive cyber defense under multi-stage attack profiles and system uncertainties. ... Development of autonomous cyber system defense strategies and action recommendations in the real-world is challenging, and includes characterizing system state uncertainties and attack-defense dynamics ... Conclusion Application of DRL methods for cyber system defense are promising, especially under dynamic adversarial uncertainties and limited system state information. ...
arXiv:2302.01595v1 fatcat:nir37j2dlrbjxhuil7dqcgazny
Deep Reinforcement Learning for Cyber Security [article]
2020 arXiv pre-print
We touch on different vital aspects, including DRL-based security methods for cyber-physical systems, autonomous intrusion detection techniques, and multi-agent DRL-based game theory simulations for defense ... The scale of Internet-connected systems has increased considerably, and these systems are being exposed to cyber attacks more than ever. ... [159] proposed the use of adversarial RL to build an autonomous defense system for SDN. ...
arXiv:1906.05799v3 fatcat:h4lujrwb5bgwngbi4xf6w347b4
Next Generation Resilient Cyber-Physical Systems [article]
2019 arXiv pre-print
In this paper, we review which requirements a CPS must meet to address the challenges of tomorrow. Two key challenges are understanding and reinforcing the resilience of CPS. ... Cyber-Physical Systems (CPS) consist of distributed engineered environments where the monitoring and surveillance tasks are governed by tightly integrated computing, communication and control technologies ... On the one hand, supervised and reinforcement learning can be used by an adversary for the purpose of system identification, an enabler for covert attacks. ...
arXiv:1907.08849v3 fatcat:ncjycpzrnnfz7hjd74g2quwvx4
Network Environment Design for Autonomous Cyberdefense [article]
2021 arXiv pre-print
Reinforcement learning (RL) has been demonstrated suitable to develop agents that play complex games with human-level performance. ... Our framework enables the development and simulation of adversaries with sophisticated behavior that includes poisoning and evasion attacks on RL network defenders. ... Introduction Since the introduction of Deep Reinforcement Learning [19] , numerous systems have successfully combined fundamental ideas from Reinforcement Learning (RL) and Deep Learning to solve hard ...
arXiv:2103.07583v1 fatcat:entwvffalvgphhpgazjtum4djq
Prospective Artificial Intelligence Approaches for Active Cyber Defence [article]
2021 arXiv pre-print
This position paper updates the roadmap for two of the most promising AI approaches -- reinforcement learning and causal inference - and describes why they could help tip the balance back towards defenders ... 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. ... provide AI-based adaptive technologies to power autonomous cyber defence systems [9] . ...
arXiv:2104.09981v1 fatcat:2jhadluy7zfi3ekjakgovfrsoi
Automated Cyber Defence: A Review [article]
2023 arXiv pre-print
Within recent times, cybercriminals have curated a variety of organised and resolute cyber attacks within a range of cyber systems, leading to consequential ramifications to private and governmental institutions ... Research within Automated Cyber Defence will allow the development and enabling intelligence response by autonomously defending networked systems through sequential decision-making agents. ... Collaboration -(G.4.1) ACO Gyms must be designed in a way to allow for multi-agent reinforcement learning (MARL) to operate -(A.4.1) Multi-Agent System representations would be required to train the automated ...
arXiv:2303.04926v1 fatcat:mv45ct6qhjhkjkverimbaiq6me
Research and Challenges of Reinforcement Learning in Cyber Defense Decision-Making for Intranet Security
2022 Algorithms
Our work provides a systematic view for understanding and solving decision-making problems in the application of reinforcement learning to cyber defense. ... Reinforcement learning has made great breakthroughs in addressing complicated decision-making problems. ... The researchers group applications of reinforcement learning in cybersecurity into DRL-based security methods for cyber-physical systems, autonomous intrusion detection techniques, and multiagent DRL-based ...
doi:10.3390/a15040134 fatcat:an3gyhnyzve6jj5r74lvqj6eki
Grand Challenges in Resilience: Autonomous System Resilience through Design and Runtime Measures [article]
2020 arXiv pre-print
For resilience-by-design, we focus on design methods in software that are needed for our cyber systems to be resilient. ... These solutions fall into two broad themes: resilience-by-design and resilience-by-reaction. We use examples of autonomous systems as the application drivers motivating cyber resilience. ... For resilience-by-design, we focus on design methods in software that are needed for our cyber systems to be resilient. ...
arXiv:1912.11598v3 fatcat:d4lf2vs4yjbbrnrg7hk65qtbbm
Reinforcement Learning for Feedback-Enabled Cyber Resilience [article]
2021 arXiv pre-print
Reinforcement Learning (RL) is an essential tool that epitomizes the feedback architectures for cyber resilience. ... A Cyber-Resilient Mechanism (CRM) adapts to the known or zero-day threats and uncertainties in real-time and strategically responds to them to maintain critical functions of the cyber systems in the event ... Reinforcement Learning in Adversarial Environment and Countermeasures As we use RL to improve the resilience of the cyber system, an intelligent attacker can also seek to compromise or mislead the RL process ...
arXiv:2107.00783v2 fatcat:faffbbapnrg5djpjhjq7rcb7ym
Symbiotic Game and Foundation Models for Cyber Deception Operations in Strategic Cyber Warfare [article]
2024 arXiv pre-print
In this landscape, cyber deception emerges as a critical component of our defense strategy against increasingly sophisticated attacks. ... FMs serve as pivotal tools across various functions for MANSCOL, including reinforcement learning, knowledge assimilation, formation of conjectures, and contextual representation. ... Meta-reinforcement learning builds on the trained models to create adaptive and self-improving cyber deception systems. ...
arXiv:2403.10570v1 fatcat:afvz6rfipffzxix7v7jdmzzrxe
« Previous Showing results 1 — 15 out of 1,904 results