Conference paper
Model Transport: Towards Scalable Transfer Learning on Manifolds
We consider the intersection of two research fields: transfer learning and statistics on manifolds. In particular, we consider, for manifold-valued data, transfer learning of tangent-space models such as Gaussians distributions, PCA, regression, or classifiers. Though one would hope to simply use ordinary Rn-transfer learning ideas, the manifold structure prevents it.
We overcome this by basing our method on inner-product-preserving parallel transport, a well-known tool widely used in other problems of statistics on manifolds in computer vision. At first, this straightforward idea seems to suffer from an obvious shortcoming: Transporting large datasets is prohibitively expensive, hindering scalability.
Fortunately, with our approach, we never transport data. Rather, we show how the statistical models themselves can be transported, and prove that for the tangent-space models above, the transport “commutes” with learning. Consequently, our compact framework, applicable to a large class of manifolds, is not restricted by the size of either the training or test sets.
We demonstrate the approach by transferring PCA and logistic-regression models of real-world data involving 3D shapes and image descriptors.
Language: | English |
---|---|
Publisher: | IEEE |
Year: | 2014 |
Pages: | 1378-1385 |
Proceedings: | 2014 IEEE Conference on Computer Vision and Pattern RecognitionIEEE Conference on Computer Vision and Pattern Recognition Workshops |
Series: | I E E E Conference on Computer Vision and Pattern Recognition. Proceedings |
ISBN: | 1479951188 , 1479951196 , 9781479951185 and 9781479951192 |
ISSN: | 2332564x and 10636919 |
Types: | Conference paper |
DOI: | 10.1109/CVPR.2014.179 |
ORCIDs: | Hauberg, Søren |
3D shapes Abt. Black Computational modeling Computer Vision Data models Manifold-Valued Data Manifolds PCA model PGA Principal component analysis R<sup>n</sup>-transfer learning ideas Riemannian Manifolds Scalable Shape Statistics on Manifolds Total quality management Transfer Learning Vectors computer vision image descriptors inner-product-preserving parallel transport learning (artificial intelligence) logistic-regression model manifold-valued data model transport principal component analysis regression analysis scalable transfer learning statistical models tangent-space models test sets training sets