Soheil Feizi Receives NSF CAREER Award

Assistant Professor Soheil Feizi receives National Science Foundation (NSF) CAREER award for a project titled "Information-Theoretic and Statistical Foundations of Generative Models".

Generative models provide a statistical understanding of data and play an important role in the success of modern machine learning in various application domains including vision, speech, natural languages, computational biology, etc. Building on the success of deep learning, recent advances in modern generative models such as Generative Adversarial Networks (GANs) and Variational AutoEncoders (VAEs) have demonstrated great potentials in revolutionizing various learning methods. Despite this progress, our understanding of some fundamental aspects of these models, required for characterizing their performance guarantees, is still in its infancy. This project aims to shed light on various statistical and computational properties of modern generative models by leveraging tools and concepts from information theory, statistics and optimization.

In addition to being an Assistant Professor of Computer Science, Feizi holds joint appointments in UMIACS and University of Maryland Center for Machine Learning. Feizi's research work focuses on understanding various fundamental and practical aspects of deep learning to develop accurate, robust, interpretable and fair machine learning methods.

The awarded project includes a comprehensive plan to integrate proposed research into an inclusive, diverse and cross-disciplinary education at the high school, undergraduate and graduate levels. One of the underlying goals of the funded project is to train the next generation of data scientist and engineers with diverse and deep expertise in machine learning, information theory, statistics, optimization and control theory. In addition to this, the award proposes efforts to mitigate the shrinking pipeline issue that concerns the decreased proportion of women in higher education especially in modern machine learning areas.

Complete Award Information-

https://www.nsf.gov/awardsearch/showAward?AWD_ID=1942230&HistoricalAwards=false

More about the NSF CAREER Awards-

https://en.wikipedia.org/wiki/National_Science_Foundation_CAREER_Awards

More about Feizi’s research-

https://www.cs.umd.edu/~sfeizi/