We consider kernel methods on general geodesic metric spaces and provide both negative and positive results. First we show that the common Gaussian kernel can only be generalized to a positive definite kernel on a geodesic metric space if the space is flat. As a result, for data on a Riemannian manifold, the geodesic Gaussian kernel is only positive definite if the Riemannian manifold is Euclidean. This implies that any attempt to design geodesic Gaussian kernels on curved Riemannian manifolds is futile. However, we show that for spaces with conditionally negative definite distances the geodesic Laplacian kernel can be generalized while retaining positive definiteness. This implies that geodesic Laplacian kernels can be generalized to some curved spaces, including spheres and hyperbolic spaces. Our theoretical results are verified empirically.
|Title of host publication||Proceedings of the 28th IEEE Conference on Computer Vision and Pattern Recognition (CVPR 2015)|
|Publication status||Published - 2015|
|Event||28th IEEE Conference on Computer Vision and Pattern Recognition - Boston, United States|
Duration: 7 Jun 2015 → 12 Jun 2015
Conference number: 28
|Conference||28th IEEE Conference on Computer Vision and Pattern Recognition|
|Period||07/06/2015 → 12/06/2015|
Bibliographical note© 2015 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including
reprinting /republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.