Current Students
Avi Schwarzschild
Avi is a Ph.D. student in the Applied Math and Scientific Computation program at the University of Maryland. He is advised by Tom Goldstein on his work in deep learning. His interest in data security and model vulnerability has led to work on adversarial attacks and data poisoning. Other projects of his focus on investigating neural networks’ ability to extrapolate from easy training tasks to more difficult problems at test time.
Manli Shu
Manli works at the intersection of machine learning and computer vision. Prior to joining our team, she completed her bachelors degree at USTC, and was an intern as Microsoft Research Asia.
Amin Ghiasi
Amin’s experience is a blend of theoretical computer science, and machine learning. His recent work is on neural architecture design, and adversarial attacks and defenses.
Zeyad Emam
Zeyad’s primary advisor is Wojciech Czaja. He has a background in harmonic analysis and image processing. His work is on improved segmentation methods for electron microscopy (work sponsored by the NIH), and he also works on reinforcement learning agents.
Chen Zhu
Chen’s interests lie in Machine Learning and applications to Computer Vision. His recent work has focused on poisoning attacks on neural networks, and defenses against these attacks. Chen received his masters degree from ShanghaiTech, and he has interned at Microsoft Research Asia, Baidu, and Intel Labs China.
Ping-Yeh Chiang
Ping-Yeh studies certifiable and provably secure methods for image classification. His focus is on improving the tightness of certifiable defenses for neural networks, and on developing new defenses against emerging threat models.
Research Scientists & Postdocs
Jonas Geiping
Jonas is a postdoctoral researcher at UMD. His background is in Mathematics, more specifically in mathematical optimization and its applications to deep learning. His current focus is on designing more secure and private ML systems, especially for federated learning, and on understanding fundamental phenomena behind generalization.
Former lab members
Ronny Huang
Ronny received his PhD in Physics at MIT in the Optics and Quantum Electronics Group under Professors Franz Kärtner and Erich Ippen at MIT. In addition to his expertise in optics and imaging, Ronny has done a lot of work in machine learning and computer vision. His recent research is focused on poisoning attacks and defenses for deep neural networks.
Parsa Saadatpanah
Parsa is a machine Learning researcher working on recommendation systems, content Discovery, and how usage data can be leveraged to create data-driven solutions. His recent work has focused on security concerns arising in content management and copyright control systems.
Micah Goldblum
Micah is completing a degree program in mathematics, with a specialization in computational methods for deep learning. His current research focuses on learning with robustness constraints in the few-shot setting, and in settings with abundant data but scarce labels. Micah previously worked on harmonic analysis and wavelet methods for imaging.
Ali Shafahi
Ali works on a range of topics related to data security for neural networks and other machine learning systems. He is particularly interested in adversarial machine learning, including evasion and poisoning attacks for deep classifiers. Before joining the group, Ali complete a PhD in Civil Engineering under the supervision of Ali Haghani, and he is an expert in operations research and transportation systems.
Zheng Xu
Zheng completed his PhD in 2019, after which he joined Google. His thesis was on optimization and machine learning. His focus was on automated and distributing optimization routines for model fitting and data science.
Soham De
Soham completed his PhD in 2018, and joined Google DeepMind in London. Soham’s thesis work included topics in machine learning and game theory. In machine learning, he worked on optimization methods, distributed algorithms, and online learning. Soham was co-advised by Dana Nau, and also collaborated with Michele Gelfand for his work on game theoretic models of human behavior and cultural bias.
Hao Li
Hao completed his PhD in 2018, and joined Amazon Research. Hao’s research interests lie at the intersection of machine learning and systems. Specifically, he is interested in designing efficient and scalable machine learning algorithms for high-performance and resource-constrained systems. Hao was co-advised by Hanan Samet.
Sohil Shah
Sohil completed his PhD in 2017, and joined Intel Research. Sohil focused on solving difficult computer vision problems using large-scale optimization, deep learning, and graphical models.