Shike Mei
Papers
1
Total Citations
368
H-Index
1
About
Shike Mei is a leading researcher in the intersection of machine learning, security, and adversarial robustness. His most influential work, "Using Machine Teaching to Identify Optimal Training-Set Attacks on Machine Learners" (2015, 368 citations), fundamentally advanced our understanding of how malicious actors can manipulate learning algorithms by contaminating training data. By framing these attacks through the lens of machine teaching, Mei provided a rigorous, optimization-based framework for identifying the most effective—and thus most dangerous—training-set poisoning strategies. This contribution has become a cornerstone in the field of adversarial machine learning, directly influencing subsequent research on data integrity, robust learning, and defensive mechanisms. His work highlights the critical vulnerabilities in modern AI systems and has shaped how the community approaches secure model deployment. With deep implications for cybersecurity, Mei's research continues to guide both theoretical and practical efforts to safeguard machine learners against sophisticated, data-driven attacks.
Research Focus
Key Achievements
Top Papers
- 1