Rising Stars in Machine Learning Talk Features Surbhi Goel from UT Austin

Subhi Goel headshot
Mon Sep 16, 2019

The University of Maryland Center for Machine Learning kicks off its Rising Stars in Machine Learning program on Monday, Sept. 23 with an hour-long talk on “Provably Efficient Algorithms for Learning Neural Networks.”

The talk will be given by Surbhi Goel, a 5th-year doctoral student in computer science from the University of Texas at Austin. Her research interests lie at the intersection of theory and machine learning, with a focus on developing efficient algorithms for training neural networks that are backed by strong theoretical guarantees.

Goel is one of three speakers chosen to present their work this fall at the University of Maryland as part of the Rising Stars in Machine Learning initiative, which supports up-and-coming female researchers as they pursue new scientific and academic opportunities.

Goel will discuss her algorithm, “Convotron,” which can be used for efficiently learning a one-hidden-layer convolutional network. The algorithm works for commonly used schemes from computer vision, including one-dimensional and two-dimensional “patch and stride” convolutions. During the second half of the talk, Goel will describe an algorithm for learning general feed forward neural networks with more than one hidden layer.

“Surbhi Goel has proposed various algorithms for shallow networks with provable guarantees, using tools from kernel methods and isotonic regression,” says Furong Huang, an assistant professor of computer science and core member of the Center for Machine Learning. “Her works have been successfully published in top machine learning conferences such as COLT, NeurIPS and ICML.”

The other Rising Stars in Machine Learning scheduled to speak this fall are Adji Bousso Dieng, a doctoral student in statistics at Columbia University, who will speak on Sept. 30; and Gintare Karolina Dziugaite, a research scientist at Element AI, who will give a talk on Oct. 21.

The UMD Center for Machine Learning, supported in part by financial and technology leader Capital One, is part of the University of Maryland Institute for Advanced Computer Studies.

Story by Colleen Curran


WHAT: Rising Stars in Machine Learning Seminar Series

WHEN: Monday, Sept. 23, from 11 a.m. to noon

WHERE: Brendan Iribe Center for Computer Science and Engineering, Room 4105