Kaidi Xu

Kaidi Xu is an assistant professor in the Department of Computer Science in Drexel's College of Computing & Informatics. His main research interest centers around analyzing security problems that happen in Artificial Intelligence (AI) technologies. He has a strong background in trustworthy machine learning, including adversarial attacks, formal robustness verification and certified defenses. In addition, he also has broad research interests in areas such as deep learning model compression and acceleration, as well as explainable AI. His research papers are published in various top conferences such as the Annual Conference on Neural Information Processing Systems (NeurIPS), International Conference on Machine Learning (ICML), International Conference on Learning Representations (ICLR), International Joint Conference on Artificial Intelligence (IJCAI), AAAI Conference on Artificial Intelligence, Conference on Computer Vision and Pattern Recognition (CVPR), European Conference on Computer Vision (ECCV), International Conference on Computational Vision (ICCV), among others. Kaidi Xu is an early researcher on Physical World Adversarial Attacks against Deep Neural Networks and is the winner of the 2nd International Verification of Neural Networks Competition (VNN-COMP’21) with the highest total score and five individual champions.