Feizi Receives Qualcomm Faculty Award

Soheil Feizi, an assistant professor of computer science with an appointment in the University of Maryland Institute for Advanced Computer Studies (UMIACS), has been selected to receive a Qualcomm Faculty Award (QFA) to advance technology used for autonomous vehicles.

The award supports key professors and their research through a charitable donation to their university.

Feizi’s research aims to leverage generative models in probabilistic predictions of time-series data. Probabilistic predictions—assigning a probability to each possible outcome—and guaranteeing the robustness of modern deep learning methods used in these predictions are critical challenges for many companies working in sensitive applications such as self-driving cars.

He is exploring the use of a modern unsupervised learning technique, known as Generative Adversarial Networks (GANs), for these problems. GANs are deep neural net architectures comprised of two nets, pitting one against the other (thus the term “adversarial”).

Since data distributions are often more complex than classical probability models and considering the empirical evidence of GAN’s success in modeling such complex distributions, Feizi believes that using GANs can provide significant gains in complex domains as an emerging technology for Qualcomm platforms and its partners.