Learning Stable Graphs from Multiple Environments with Selection Bias

Yue He; Peng Cui; Jianxin Ma; Hao Zou; Xiaowei Wang; Hongxia Yang; Philip S. Yu

doi:10.1145/3394486.3403270

Learning Stable Graphs from Multiple Environments with Selection Bias

Yue He, Peng Cui, Jianxin Ma, Hao Zou, Xiaowei Wang, Hongxia Yang, Philip S. Yu

Research output: Chapter in book / Conference proceeding › Conference article published in proceeding or book › Academic research › peer-review

11 Citations (Scopus)

Abstract

Nowadays graph has become a general and powerful representation to describe the rich relationships among different kinds of entities via the underlying patterns encoded in its structure. The knowledge (more generally) accumulated in graph is expected to be able to cross populations from one to another and the past to future. However the data collection process of graph generation is full of known or unknown sample selection biases, leading to spurious correlations among entities, especially in the non-stationary and heterogeneous environments. In this paper, we target the problem of learning stable graphs from multiple environments with selection bias. We purpose a Stable Graph Learning (SGL) framework to learn a graph that can capture general relational patterns which are irrelevant with the selection bias in an unsupervised way. Extensive experimental results from both simulation and real data demonstrate that our method could significantly benefit the generalization capacity of graph structure.

Original language	English
Title of host publication	KDD 2020 - Proceedings of the 26th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining
Publisher	Association for Computing Machinery
Pages	2194-2202
Number of pages	9
ISBN (Electronic)	9781450379984
DOIs	https://doi.org/10.1145/3394486.3403270
Publication status	Published - 23 Aug 2020
Externally published	Yes
Event	26th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, KDD 2020 - Virtual, Online, United States Duration: 23 Aug 2020 → 27 Aug 2020

Publication series

Name	Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining

Conference

Conference	26th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, KDD 2020
Country/Territory	United States
City	Virtual, Online
Period	23/08/20 → 27/08/20

Keywords

graph structure
multiple environments
selection bias
stability

ASJC Scopus subject areas

Software
Information Systems

More information

10.1145/3394486.3403270

Cite this

He, Y., Cui, P., Ma, J., Zou, H., Wang, X., Yang, H., & Yu, P. S. (2020). Learning Stable Graphs from Multiple Environments with Selection Bias. In KDD 2020 - Proceedings of the 26th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (pp. 2194-2202). (Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining). Association for Computing Machinery. https://doi.org/10.1145/3394486.3403270

He, Yue ; Cui, Peng ; Ma, Jianxin et al. / Learning Stable Graphs from Multiple Environments with Selection Bias. KDD 2020 - Proceedings of the 26th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. Association for Computing Machinery, 2020. pp. 2194-2202 (Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining).

@inproceedings{d45e1becfde54e4bbbf3fe60ca4d9c87,

title = "Learning Stable Graphs from Multiple Environments with Selection Bias",

abstract = "Nowadays graph has become a general and powerful representation to describe the rich relationships among different kinds of entities via the underlying patterns encoded in its structure. The knowledge (more generally) accumulated in graph is expected to be able to cross populations from one to another and the past to future. However the data collection process of graph generation is full of known or unknown sample selection biases, leading to spurious correlations among entities, especially in the non-stationary and heterogeneous environments. In this paper, we target the problem of learning stable graphs from multiple environments with selection bias. We purpose a Stable Graph Learning (SGL) framework to learn a graph that can capture general relational patterns which are irrelevant with the selection bias in an unsupervised way. Extensive experimental results from both simulation and real data demonstrate that our method could significantly benefit the generalization capacity of graph structure.",

keywords = "graph structure, multiple environments, selection bias, stability",

author = "Yue He and Peng Cui and Jianxin Ma and Hao Zou and Xiaowei Wang and Hongxia Yang and Yu, \{Philip S.\}",

note = "Publisher Copyright: {\textcopyright} 2020 ACM.; 26th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, KDD 2020 ; Conference date: 23-08-2020 Through 27-08-2020",

year = "2020",

month = aug,

day = "23",

doi = "10.1145/3394486.3403270",

language = "English",

series = "Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining",

publisher = "Association for Computing Machinery",

pages = "2194--2202",

booktitle = "KDD 2020 - Proceedings of the 26th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining",

}

He, Y, Cui, P, Ma, J, Zou, H, Wang, X, Yang, H & Yu, PS 2020, Learning Stable Graphs from Multiple Environments with Selection Bias. in KDD 2020 - Proceedings of the 26th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, Association for Computing Machinery, pp. 2194-2202, 26th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, KDD 2020, Virtual, Online, United States, 23/08/20. https://doi.org/10.1145/3394486.3403270

Learning Stable Graphs from Multiple Environments with Selection Bias. / He, Yue; Cui, Peng; Ma, Jianxin et al.
KDD 2020 - Proceedings of the 26th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. Association for Computing Machinery, 2020. p. 2194-2202 (Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining).

Research output: Chapter in book / Conference proceeding › Conference article published in proceeding or book › Academic research › peer-review

TY - GEN

T1 - Learning Stable Graphs from Multiple Environments with Selection Bias

AU - He, Yue

AU - Cui, Peng

AU - Ma, Jianxin

AU - Zou, Hao

AU - Wang, Xiaowei

AU - Yang, Hongxia

AU - Yu, Philip S.

PY - 2020/8/23

Y1 - 2020/8/23

N2 - Nowadays graph has become a general and powerful representation to describe the rich relationships among different kinds of entities via the underlying patterns encoded in its structure. The knowledge (more generally) accumulated in graph is expected to be able to cross populations from one to another and the past to future. However the data collection process of graph generation is full of known or unknown sample selection biases, leading to spurious correlations among entities, especially in the non-stationary and heterogeneous environments. In this paper, we target the problem of learning stable graphs from multiple environments with selection bias. We purpose a Stable Graph Learning (SGL) framework to learn a graph that can capture general relational patterns which are irrelevant with the selection bias in an unsupervised way. Extensive experimental results from both simulation and real data demonstrate that our method could significantly benefit the generalization capacity of graph structure.

AB - Nowadays graph has become a general and powerful representation to describe the rich relationships among different kinds of entities via the underlying patterns encoded in its structure. The knowledge (more generally) accumulated in graph is expected to be able to cross populations from one to another and the past to future. However the data collection process of graph generation is full of known or unknown sample selection biases, leading to spurious correlations among entities, especially in the non-stationary and heterogeneous environments. In this paper, we target the problem of learning stable graphs from multiple environments with selection bias. We purpose a Stable Graph Learning (SGL) framework to learn a graph that can capture general relational patterns which are irrelevant with the selection bias in an unsupervised way. Extensive experimental results from both simulation and real data demonstrate that our method could significantly benefit the generalization capacity of graph structure.

KW - graph structure

KW - multiple environments

KW - selection bias

KW - stability

UR - https://www.scopus.com/pages/publications/85090404481

U2 - 10.1145/3394486.3403270

DO - 10.1145/3394486.3403270

M3 - Conference article published in proceeding or book

AN - SCOPUS:85090404481

T3 - Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining

SP - 2194

EP - 2202

BT - KDD 2020 - Proceedings of the 26th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining

PB - Association for Computing Machinery

T2 - 26th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, KDD 2020

Y2 - 23 August 2020 through 27 August 2020

ER -

He Y, Cui P, Ma J, Zou H, Wang X, Yang H et al. Learning Stable Graphs from Multiple Environments with Selection Bias. In KDD 2020 - Proceedings of the 26th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. Association for Computing Machinery. 2020. p. 2194-2202. (Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining). doi: 10.1145/3394486.3403270

Learning Stable Graphs from Multiple Environments with Selection Bias

Abstract

Publication series

Conference

Keywords

ASJC Scopus subject areas

More information

Other files and links

Fingerprint

Cite this