Home >

news Help

Publication Information


Title
Japanese: 
English:Learning Graph Neural Networks with Noisy Labels 
Author
Japanese: Nguyen Hoang Thai, Choong Jun Jin, 村田剛志.  
English: Hoang Nguyen, Jun Jin Choong, Tsuyoshi MURATA.  
Language English 
Journal/Book name
Japanese: 
English: 
Volume, Number, Page         pp. 1-5
Published date May 6, 2019 
Publisher
Japanese: 
English: 
Conference name
Japanese: 
English:The 2nd Learning from Limited Labeled Data (LLD) Workshop 
Conference site
Japanese: 
English:New Orleans 
Official URL https://openreview.net/forum?id=r1xOmNmxuN
 
Abstract We study the robustness to symmetric label noise of GNNs training procedures. By combining the nonlinear neural message-passing models (e.g. Graph Isomorphism Networks, GraphSAGE, etc.) with loss correction methods, we present a noise-tolerant approach for the graph classification task. Our experiments show that test accuracy can be improved under the artificial symmetric noisy setting.

©2007 Tokyo Institute of Technology All rights reserved.