12:00 arrival and lunch served
12:30 talk starts
13:30 end of Q&A
Title: Unsupervised Metric Learning: Diffusion, Ensemble, and Subspaces
In this talk, we will discuss a series of diffusion-based metric learning algorithms, which utilize the intrinsic data manifold without the time-consuming process for the manifold reconstruction. Using single or multiple input metrics (fusion), we will see significantly enhanced state-of-the-art results in applications for retrieval, classification, clustering, and segmentation. The diffusion idea can be further carried to construct k-NN graphs for large-scale high-dimensional data by excising three ideas altogether: ensemble, divide-and-conquer, and subspace learning. We show large speed-up in dealing with large scale datasets.
Zhuowen Tu is an assistant professor in the Department of Cognitive Science, and the Department of Computer Science and Engineering, University of California, San Diego. Before joining UCSD, he was an assistant professor at UCLA. Between 2011 and 2013, he took a leave to work at Microsoft Research Asia. He received his Ph.D. from the Ohio State University and his M.E. from Tsinghua University.
Please bring your tickets for validation.