DC Field | Value | Language |
---|---|---|
dc.contributor.author | Kim, Jongmin | ko |
dc.contributor.author | Kim, Saesup | ko |
dc.contributor.author | Kim, Sungwoong | ko |
dc.contributor.author | Yoo, Chang-Dong | ko |
dc.date.accessioned | 2019-11-28T07:20:10Z | - |
dc.date.available | 2019-11-28T07:20:10Z | - |
dc.date.created | 2019-11-27 | - |
dc.date.created | 2019-11-27 | - |
dc.date.created | 2019-11-27 | - |
dc.date.created | 2019-11-27 | - |
dc.date.created | 2019-11-27 | - |
dc.date.issued | 2019-06-18 | - |
dc.identifier.citation | IEEE Conference on Computer Vision and Pattern Recognition, pp.11 - 20 | - |
dc.identifier.issn | 1063-6919 | - |
dc.identifier.uri | http://hdl.handle.net/10203/268687 | - |
dc.description.abstract | In this paper, we propose a novel edge-labeling graph neural network (EGNN), which adapts a deep neural network on the edge-labeling graph, for few-shot learning. The previous graph neural network (GNN) approaches in few-shot learning have been based on the node-labeling framework, which implicitly models the intra-cluster similarity and the inter-cluster dissimilarity. In contrast, the proposed EGNN learns to predict the edge-labels rather than the node-labels on the graph that enables the evolution of an explicit clustering by iteratively updating the edge-labels with direct exploitation of both intra-cluster similarity and the inter-cluster dissimilarity. It is also well suited for performing on various numbers of classes without retraining, and can be easily extended to perform a transductive inference. The parameters of the EGNN are learned by episodic training with an edge-labeling loss to obtain a well-generalizable model for unseen low-data problem. On both of the supervised and semi-supervised few-shot image classification tasks with two benchmark datasets, the proposed EGNN significantly improves the performances over the existing GNNs. | - |
dc.language | English | - |
dc.publisher | Computer Vision Foundation / IEEE Computer Society | - |
dc.title | Edge-Labeling Graph Neural Network for Few-shot Learning | - |
dc.type | Conference | - |
dc.identifier.wosid | 000529484000002 | - |
dc.identifier.scopusid | 2-s2.0-85078772649 | - |
dc.type.rims | CONF | - |
dc.citation.beginningpage | 11 | - |
dc.citation.endingpage | 20 | - |
dc.citation.publicationname | IEEE Conference on Computer Vision and Pattern Recognition | - |
dc.identifier.conferencecountry | US | - |
dc.identifier.conferencelocation | Long Beach, CA | - |
dc.identifier.doi | 10.1109/CVPR.2019.00010 | - |
dc.contributor.localauthor | Yoo, Chang-Dong | - |
dc.contributor.nonIdAuthor | Kim, Saesup | - |
dc.contributor.nonIdAuthor | Kim, Sungwoong | - |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.