Skip to content

Computer Science Department Colloquium Series Presents Yingyu Liang


“Modern Machine Learning: New Challenges and Foundations”

 Yingyu Liang

Princeton University


Machine learning has recently achieved great empirical success. This comes along with new challenges, such as sophisticated models that lack rigorous analysis, simple algorithms with practical success on hard optimization problems, and handling large scale datasets under resource constraints. In this talk, I will present some of my work in addressing such challenges.

This first part of the talk focuses on learning semantic representations for text data. Recent advances in natural language processing build upon the approach of embedding words as low dimensional vectors. The fundamental observation that empirically justifies this approach is that these vectors can capture semantic relations. A probabilistic model for generating text is proposed to mathematically explain this observation and existing popular embedding algorithms. It also reveals surprising connections to classical notions such as Pointwise Mutual Information, and allows to design novel, simple, and practical algorithms for applications such as sentence embedding.

I will then describe recent progress in analyzing the loss surface of training one-hidden-layer neural networks. Finally, I will briefly describe a simple algorithm for clustering large-scale data distributed over different locations, which has provable guarantees on solution quality, nearly optimal communication cost, and strong empirical performance.


Yingyu Liang is an associate research scholar in the Computer Science Department at Princeton University. His research interests are providing rigorous analysis for machine learning models and designing efficient algorithms for applications. He received an M.S. in Computer Science from Tsinghua University in 2010, and a Ph.D. degree from Georgia Institute of Technology in 2014. He was a postdoctoral fellow in 2014-2015 and a lecturer in 2015-2016 in the Computer Science Department at Princeton University.

         Date:       Thursday, February 16th, 2017

         Time:       11:30am to 12:30pm

         Location:  ECS South 2.102 TI Auditorium

Refreshments will be served at 11:15am