Improved Local Coordinate Coding using Local Tangents

9 years 1 months ago
Improved Local Coordinate Coding using Local Tangents
Local Coordinate Coding (LCC), introduced in (Yu et al., 2009), is a high dimensional nonlinear learning method that explicitly takes advantage of the geometric structure of the data. Its successful use in the winning system of last year's Pascal image classification Challenge (Everingham, 2009) shows that the ability to integrate geometric information is critical for some real world machine learning applications. This paper further develops the idea of integrating geometry in machine learning by extending the original LCC method to include local tangent directions. These new correction terms lead to better approximation of high dimensional nonlinear functions when the underlying data manifold is locally relatively flat. The method significantly reduces the number of anchor points needed in LCC, which not only reduces computational cost, but also improves prediction performance. Experiments are included to demonstrate that this method is more effective than the original LCC metho...
Kai Yu, Tong Zhang
Added 09 Nov 2010
Updated 09 Nov 2010
Type Conference
Year 2010
Where ICML
Authors Kai Yu, Tong Zhang
Comments (0)