dc.citation.conferencePlace |
UK |
- |
dc.citation.conferencePlace |
University of Bath |
- |
dc.citation.endPage |
729 |
- |
dc.citation.startPage |
713 |
- |
dc.citation.title |
European Conference on Computer Vision |
- |
dc.contributor.author |
Kim, Kwang In |
- |
dc.date.accessioned |
2023-12-19T20:07:25Z |
- |
dc.date.available |
2023-12-19T20:07:25Z |
- |
dc.date.created |
2019-02-28 |
- |
dc.date.issued |
2016-10-11 |
- |
dc.description.abstract |
We observe the distances between estimated function outputs on data points to create an anisotropic graph Laplacian which, through an iterative process, can itself be regularized. Our algorithm is instantiated as a discrete regularizer on a graph’s diffusivity operator. This idea is grounded in the theory that regularizing the diffusivity operator corresponds to regularizing the metric on Riemannian manifolds, which further corresponds to regularizing the anisotropic Laplace-Beltrami operator. We show that our discrete regularization framework is consistent in the sense that it converges to (continuous) regularization on underlying data generating manifolds. In semi-supervised learning experiments, across ten standard datasets, our diffusion of Laplacian approach has the lowest average error rate of eight different established and stateof- the-art approaches, which shows the promise of our approach. © Springer International Publishing AG 2016. |
- |
dc.identifier.bibliographicCitation |
European Conference on Computer Vision, pp.713 - 729 |
- |
dc.identifier.doi |
10.1007/978-3-319-46454-1_43 |
- |
dc.identifier.issn |
0302-9743 |
- |
dc.identifier.scopusid |
2-s2.0-84990067661 |
- |
dc.identifier.uri |
https://scholarworks.unist.ac.kr/handle/201301/32617 |
- |
dc.identifier.url |
https://link.springer.com/chapter/10.1007%2F978-3-319-46454-1_43 |
- |
dc.language |
영어 |
- |
dc.publisher |
ECCV 2016 |
- |
dc.title |
Semi-supervised learning based on joint diffusion of graph functions and Laplacians |
- |
dc.type |
Conference Paper |
- |
dc.date.conferenceDate |
2016-10-11 |
- |