Cross-Space Active Learning on Graph Convolutional Networks

Yufei Tao, Hao Wu, Shiyuan Deng
Proceedings of the 39th International Conference on Machine Learning, PMLR 162:21133-21145, 2022.

Abstract

This paper formalizes cross-space active learning on a graph convolutional network (GCN). The objective is to attain the most accurate hypothesis available in any of the instance spaces generated by the GCN. Subject to the objective, the challenge is to minimize the label cost, measured in the number of vertices whose labels are requested. Our study covers both budget algorithms which terminate after a designated number of label requests, and verifiable algorithms which terminate only after having found an accurate hypothesis. A new separation in label complexity between the two algorithm types is established. The separation is unique to GCNs.

Cite this Paper

BibTeX
@InProceedings{pmlr-v162-tao22a, title = {Cross-Space Active Learning on Graph Convolutional Networks}, author = {Tao, Yufei and Wu, Hao and Deng, Shiyuan}, booktitle = {Proceedings of the 39th International Conference on Machine Learning}, pages = {21133--21145}, year = {2022}, editor = {Chaudhuri, Kamalika and Jegelka, Stefanie and Song, Le and Szepesvari, Csaba and Niu, Gang and Sabato, Sivan}, volume = {162}, series = {Proceedings of Machine Learning Research}, month = {17--23 Jul}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v162/tao22a/tao22a.pdf}, url = {https://proceedings.mlr.press/v162/tao22a.html}, abstract = {This paper formalizes cross-space active learning on a graph convolutional network (GCN). The objective is to attain the most accurate hypothesis available in any of the instance spaces generated by the GCN. Subject to the objective, the challenge is to minimize the label cost, measured in the number of vertices whose labels are requested. Our study covers both budget algorithms which terminate after a designated number of label requests, and verifiable algorithms which terminate only after having found an accurate hypothesis. A new separation in label complexity between the two algorithm types is established. The separation is unique to GCNs.} }
Endnote
%0 Conference Paper %T Cross-Space Active Learning on Graph Convolutional Networks %A Yufei Tao %A Hao Wu %A Shiyuan Deng %B Proceedings of the 39th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2022 %E Kamalika Chaudhuri %E Stefanie Jegelka %E Le Song %E Csaba Szepesvari %E Gang Niu %E Sivan Sabato %F pmlr-v162-tao22a %I PMLR %P 21133--21145 %U https://proceedings.mlr.press/v162/tao22a.html %V 162 %X This paper formalizes cross-space active learning on a graph convolutional network (GCN). The objective is to attain the most accurate hypothesis available in any of the instance spaces generated by the GCN. Subject to the objective, the challenge is to minimize the label cost, measured in the number of vertices whose labels are requested. Our study covers both budget algorithms which terminate after a designated number of label requests, and verifiable algorithms which terminate only after having found an accurate hypothesis. A new separation in label complexity between the two algorithm types is established. The separation is unique to GCNs.
APA
Tao, Y., Wu, H. & Deng, S.. (2022). Cross-Space Active Learning on Graph Convolutional Networks. Proceedings of the 39th International Conference on Machine Learning, in Proceedings of Machine Learning Research 162:21133-21145 Available from https://proceedings.mlr.press/v162/tao22a.html.

Related Material