Document detail
ID
oai:arXiv.org:2311.00859
Topic
Computer Science - Machine Learnin... Computer Science - Artificial Inte... Computer Science - Cryptography an... Computer Science - Multiagent Syst...Author
Lu, Ziqing Liu, Guanlin Lai, Lifeng Xu, WeiyuCategory
Computer Science
Year
2023
listing date
3/6/2024
Keywords
optimalAbstract
Finding optimal adversarial attack strategies is an important topic in reinforcement learning and the Markov decision process.
Previous studies usually assume one all-knowing coordinator (attacker) for whom attacking different recipient (victim) agents incurs uniform costs.
However, in reality, instead of using one limitless central attacker, the attacks often need to be performed by distributed attack agents.
We formulate the problem of performing optimal adversarial agent-to-agent attacks using distributed attack agents, in which we impose distinct cost constraints on each different attacker-victim pair.
We propose an optimal method integrating within-step static constrained attack-resource allocation optimization and between-step dynamic programming to achieve the optimal adversarial attack in a multi-agent system.
Our numerical results show that the proposed attacks can significantly reduce the rewards received by the attacked agents.
;Comment: Submitted to ICCASP2024
Lu, Ziqing,Liu, Guanlin,Lai, Lifeng,Xu, Weiyu, 2023, Optimal Cost Constrained Adversarial Attacks For Multiple Agent Systems
Content-Adaptive Variable Framerate Encoding Scheme for Green Live Streaming
streaming compression ladder yields representation cvfr-eco efficiency vmaf optimized average energy bitrate
Is Neurodevelopmental Assessment in Early Childhood Predictive of Performance Assessed Later in Childhood and Adolescence in Sub-Saharan Africa? A Systematic Review of the Literature
sub-saharan performance review systematic africa age childhood
