Views Can Be Deceiving: Improved SSL Through Feature Space Augmentation
Kimia Hamidieh
Kimia is a PhD student at University of Toronto and Vector Institute visiting MIT. Her research focuses on understanding how self-supervised pre-training strategies represent data to build models that generalize well out-of-distribution, and developing methods that enable efficient and reliable adaptation. She is also interested in leveraging properties of large models for reasoning and robustness to distribution shifts.
Haoran Zhang
Haoran is a third year PhD student in EECS at MIT. He is generally interested in building robust machine learning models that maintain their performance and fairness across out-of-distribution environments, as well as applying such models to the healthcare setting. Haoran previously received his M.Sc. at the University of Toronto under the co-supervision of Dr. Marzyeh Ghassemi and Dr. Quaid Morris, and his B.Eng. from McMaster University.