Mei wins Best Paper at ICML

UMSI Associate Professor Qiaozhu Mei and his co-authors have been awarded the Best Paper prize out of more than 1,000 submissions at the International Conference on Machine Learning (ICML), held June 21-26 in Beijing, China. The award was announced at the conference on June 24.

The paper, “Understanding the Limiting Factors of Topic Modeling via Posterior Contraction Analysis,” provides practical guidance for researchers on how to identify suitable data sets for topic models and how to specify particular model parameters. Authors on the paper, in addition to Mei, are Jian Tang and Ming Zhang of Peking University) and Zhaoshi Meng and  Xuanlong Nguyen at the University of Michigan. Jian Tang worked on the research while at UMSI as a visiting scholar.

ICML is one of the top conferences of machine learning and this year received over 1,000 submissions.

Abstract

Paper

Posted on June 24, 2014