Misha Khodak
Position title: Assistant Professor
Email: khodak@wisc.edu
Address:
Morgridge Hall 5512
Misha Khodak
I am an assistant professor of computer sciences at the University of Wisconsin-Madison, where I study foundations and applications of machine learning. Most recently, I have been working on specialized foundation models and on incorporating AI tools into algorithm design and scientific computing.
My body of work includes fundamental theory for modern meta-learning (scalable methods that “learn-to-learn” using multiple learning tasks as data) and end-to-end guarantees for learning-augmented algorithms (algorithms that incorporate learned predictions about their instances to improve performance). These results are based on a set of theoretical tools that port the idea of surrogate upper bounds from supervised learning to learning algorithmic cost functions. In addition to providing natural measures of task-similarity, this approach often yields effective and practical methods, such as for personalized federated learning and scientific computing. I have also led the push to develop automated ML methods for diverse tasks and have worked on efficient deep learning, neural architecture search, and natural language processing.
Previously, I spent a year as a fellow at Princeton Language & Intelligence after completing a PhD in CS at CMU.
To get in touch, email me at khodak@wisc.edu. Please read my FAQ first if you are a prospective advisee or student.