Information Transfer across Classes: Semi-supervised Zero-Shot Learning
Yuhong Guo
Assistant Professor
Department of Computer and Information Sciences, Temple University
Abstract: Even in the era of Big Data, the need for manually annotated data remains a critical bottleneck in developing automated prediction systems. Due to the dramatic expanse of data categories and the lack of labeled instances, zero-shot learning, which transfers knowledge from observed classes to recognize unseen classes, has started drawing a lot of attention from the research community. In this talk, I will present a semi-supervised max-margin learning framework that integrates the semi-supervised classification problem over observed classes and the unsupervised clustering problem over unseen classes together to tackle zero-shot multi-class classification. This proposed framework can also conveniently incorporate auxiliary label semantic knowledge to improve zero-shot learning. The experimental results on three standard image data sets demonstrate the efficacy of the proposed framework.
Bio: Dr. Guo, Yuhong received her PhD from the University of Alberta in 2007. She has been a Research Fellow at the Australian National University and is currently an Assistant Professor in the Department of Computer and Information Sciences at Temple University. Her primary research area is machine learning, with applications in natural language processing, bioinformatics and computer vision. Dr. Guo has published about sixty refereed papers in these areas, including NIPS, ICML, ECML, AAAI, IJCAI, UAI, AISTATS, CVPR, EMNLP, COLING and CoNLL. She has also received a number of awards for her research, including best paper prizes at IJCAI and AAAI.