Resilient distributed optimization under Adversarial Attacks

This research examines resilient distributed optimization (RDO) effects of multi-agent systems against adversary attacks. We develop RDO algorithms for trusted agents against adversary attacks. The research's priority is to understand and exploit vulnerabilities of the current RDO of multi-agent systems in the present of cyber-threats. We study exploited vulnerability patterns in RDO algorithms to determine the strategy of unauthorized users or adversaries in compromising data confidentiality, data integrity, and data availability. Hopefully, we will gain insight into that adversary's attacking strategy if we can establish vulnerability exploitation generalizability in the trusted agent systems.