2019 - 2020

1. Adversarial Machine Learning

• Exploring the vulnerability of machine learning mdoels to adverarial attacks under both black-box and white-box settings.

• Defense machine learning models against adversarial attacks.

• Improving adversarial training as a defense technique for model robustness.

• Model interpretation and understanding.

• Image, video, text, audio adversarial examples.

• From Lp-bounded to unrestricted adversarial perturbations.

2. Robust Supervised/Weakly-Supervised Learning

• Learning with noisy/adversarial labels.

• Learning with input noise.

• Learning with adversarial back-door/trojan attacks.

• Learning in noisy and dynamic real-world environments.