Unist Safe AI Lab

 

Welcome to Safe AI Lab at Department of Industrial Engineering, UNIST

Safe AI Lab specializes in the security of AI models, focusing on privacy-preserving machine learning, fairness verification and certification, and security-enhanced machine learning. We are interested in inherent vulnerabilities within AI models.

Privacy-preserving machine learning: secure computations and differential privacy

Fairness-aware machine learning: fairness verification and certification

Security-enhanced machine learning: adversarial robustness, poisoning attacks and model extraction attacks

Deep representation learning

Transfer learning

 

 

50, UNIST-gil, Eonyang-eup, Ulju-gun, Ulsan, Republic of Korea

☎️ +82 52 217 7599

✉️ srompark@unist.ac.kr