Yuhong Guo, School of Computer Science
Carleton University
The need for annotated data is a fundamental bottleneck in developing automated prediction systems. A key strategy for reducing the reliance on human annotation is to exploit generalized information transfer, where a limited data resource is augmented with labeled data collected from related sources.
In this talk, I will first introduce the problem of generalized information transfer, then discuss a special form of information transfer, zero-shot learning (ZSL), where knowledge is transferred between classes. For the ZSL problem, I will present a learning methodology based on discriminative non-negative matrix factorization that provides an effective approach for recognizing novel object categories in images. I will also discuss extensions that use end-to-end deep learning to achieve enhanced ZSL performance. The talk will conclude with an overview of my other work on reducing the reliance on manual annotation in machine learning.
Bio: Dr. Yuhong Guo is an Associate Professor at Carleton University and a Canada Research Chair in Machine Learning. She received her PhD from the University of Alberta, and has previously worked at the Australian National University and Temple University.
Her research interests include machine learning, artificial intelligence, computer vision, and natural language processing. Dr. Guo has published over eighty refereed papers in these areas and received paper awards from both IJCAI and AAAI. She has served as an Area Chair for AAAI, IJCAI and ACML, a reviewer for NeurIPS and ICML, and is currently serving as an Associate Editor for TPAMI.