BIRS Workshop Lecture Videos

Banff International Research Station Logo

BIRS Workshop Lecture Videos

What Is Beyond Sparse Coding? Wu, Yingnian

Description

Many types of data such as natural images admit sparse representations by redundant dictionaries of basis functions (or regressors), and these dictionaries can either be designed or learned from training data. However, it is still unclear how to go beyond sparsity and continue to learn structures behind the sparse representations. In this talk, I shall review some recent progresses and the major issues and difficulties that need to be addressed. I shall also present our own recent work that seeks to learn dictionaries of compositional patterns in the sparse representations. Based on joint work with Jianwen Xie, Wenze Hu and Song-Chun Zhu.

Item Media

Item Citations and Data

Rights

Attribution-NonCommercial-NoDerivs 2.5 Canada