Behind Distribution Shift: Mining Driving Forces of Changes and Causal Arrows

Biwei HUANG; Kun ZHANG; Jiji ZHANG; Ruben SANCHEZ-ROMERO; Clark GLYMOUR; Bernhard SCHÖKOPF

doi:10.1109/ICDM.2017.114

Behind Distribution Shift: Mining Driving Forces of Changes and Causal Arrows

Biwei HUANG, Kun ZHANG, Jiji ZHANG, Ruben SANCHEZ-ROMERO, Clark GLYMOUR, Bernhard SCHÖKOPF

Department of Philosophy

Research output: Book Chapters | Papers in Conference Proceedings › Conference paper (refereed) › Research › peer-review

26 Citations (Scopus)

Abstract

We address two important issues in causal discovery from nonstationary or heterogeneous data, where parameters associated with a causal structure may change over time or across data sets. First, we investigate how to efficiently estimate the "driving force" of the nonstationarity of a causal mechanism. That is, given a causal mechanism that varies over time or across data sets and whose qualitative structure is known, we aim to extract from data a low-dimensional and interpretable representation of the main components of the changes. For this purpose we develop a novel kernel embedding of nonstationary conditional distributions that does not rely on sliding windows. Second, the embedding also leads to a measure of dependence between the changes of causal modules that can be used to determine the directions of many causal arrows. We demonstrate the power of our methods with experiments on both synthetic and real data.

Original language	English
Title of host publication	Proceedings : 17th IEEE International Conference on Data Mining, ICDM 2017
Editors	George KARYPIS, Srinivas ALU, Vijay RAGHAVAN, Xindong WU, Lucio MIELE
Publisher	IEEE
Pages	913-918
Number of pages	6
ISBN (Electronic)	9781538638347
DOIs	https://doi.org/10.1109/ICDM.2017.114
Publication status	Published - 2017
Event	IEEE International Conference on Data Mining - United States, New Orleans, United States Duration: 18 Nov 2017 → 21 Nov 2017

Publication series

Name	Proceedings - IEEE International Conference on Data Mining, ICDM
Publisher	IEEE
ISSN (Print)	1550-4786
ISSN (Electronic)	2374-8486

Conference

Conference	IEEE International Conference on Data Mining
Country/Territory	United States
City	New Orleans
Period	18/11/17 → 21/11/17
Other	IEEE computer society

Bibliographical note

Publisher Copyright:
© 2017 IEEE.

Funding

This project was supported by the National Institutes of Health (NIH) under Award Numbers NIH–1R01EB022858-01 FAIN–R01EB022858, NIH–1R01LM012087, and NIH– 5U54HG008540-02 FAIN–U54HG008540.

Keywords

Causal discovery
Data distribution shift
Kernel mean embedding
Nonstationary driving force

Access to Document

10.1109/ICDM.2017.114

Cite this

HUANG, B., ZHANG, K., ZHANG, J., SANCHEZ-ROMERO, R., GLYMOUR, C., & SCHÖKOPF, B. (2017). Behind Distribution Shift: Mining Driving Forces of Changes and Causal Arrows. In G. KARYPIS, S. ALU, V. RAGHAVAN, X. WU, & L. MIELE (Eds.), Proceedings : 17th IEEE International Conference on Data Mining, ICDM 2017 (pp. 913-918). (Proceedings - IEEE International Conference on Data Mining, ICDM). IEEE. https://doi.org/10.1109/ICDM.2017.114

HUANG, Biwei ; ZHANG, Kun ; ZHANG, Jiji et al. / Behind Distribution Shift: Mining Driving Forces of Changes and Causal Arrows. Proceedings : 17th IEEE International Conference on Data Mining, ICDM 2017. editor / George KARYPIS ; Srinivas ALU ; Vijay RAGHAVAN ; Xindong WU ; Lucio MIELE. IEEE, 2017. pp. 913-918 (Proceedings - IEEE International Conference on Data Mining, ICDM).

@inproceedings{3a04a75801df4f459e86899d9999bc98,

title = "Behind Distribution Shift: Mining Driving Forces of Changes and Causal Arrows",

abstract = "We address two important issues in causal discovery from nonstationary or heterogeneous data, where parameters associated with a causal structure may change over time or across data sets. First, we investigate how to efficiently estimate the {"}driving force{"} of the nonstationarity of a causal mechanism. That is, given a causal mechanism that varies over time or across data sets and whose qualitative structure is known, we aim to extract from data a low-dimensional and interpretable representation of the main components of the changes. For this purpose we develop a novel kernel embedding of nonstationary conditional distributions that does not rely on sliding windows. Second, the embedding also leads to a measure of dependence between the changes of causal modules that can be used to determine the directions of many causal arrows. We demonstrate the power of our methods with experiments on both synthetic and real data.",

keywords = "Causal discovery, Data distribution shift, Kernel mean embedding, Nonstationary driving force",

author = "Biwei HUANG and Kun ZHANG and Jiji ZHANG and Ruben SANCHEZ-ROMERO and Clark GLYMOUR and Bernhard SCH{\"O}KOPF",

note = "Publisher Copyright: {\textcopyright} 2017 IEEE.; IEEE International Conference on Data Mining ; Conference date: 18-11-2017 Through 21-11-2017",

year = "2017",

doi = "10.1109/ICDM.2017.114",

language = "English",

series = "Proceedings - IEEE International Conference on Data Mining, ICDM",

publisher = "IEEE",

pages = "913--918",

editor = "George KARYPIS and Srinivas ALU and Vijay RAGHAVAN and Xindong WU and Lucio MIELE",

booktitle = "Proceedings : 17th IEEE International Conference on Data Mining, ICDM 2017",

}

HUANG, B, ZHANG, K, ZHANG, J, SANCHEZ-ROMERO, R, GLYMOUR, C & SCHÖKOPF, B 2017, Behind Distribution Shift: Mining Driving Forces of Changes and Causal Arrows. in G KARYPIS, S ALU, V RAGHAVAN, X WU & L MIELE (eds), Proceedings : 17th IEEE International Conference on Data Mining, ICDM 2017. Proceedings - IEEE International Conference on Data Mining, ICDM, IEEE, pp. 913-918, IEEE International Conference on Data Mining, New Orleans, United States, 18/11/17. https://doi.org/10.1109/ICDM.2017.114

Behind Distribution Shift: Mining Driving Forces of Changes and Causal Arrows. / HUANG, Biwei; ZHANG, Kun; ZHANG, Jiji et al.
Proceedings : 17th IEEE International Conference on Data Mining, ICDM 2017. ed. / George KARYPIS; Srinivas ALU; Vijay RAGHAVAN; Xindong WU; Lucio MIELE. IEEE, 2017. p. 913-918 (Proceedings - IEEE International Conference on Data Mining, ICDM).

Research output: Book Chapters | Papers in Conference Proceedings › Conference paper (refereed) › Research › peer-review

TY - GEN

T1 - Behind Distribution Shift: Mining Driving Forces of Changes and Causal Arrows

AU - HUANG, Biwei

AU - ZHANG, Kun

AU - ZHANG, Jiji

AU - SANCHEZ-ROMERO, Ruben

AU - GLYMOUR, Clark

AU - SCHÖKOPF, Bernhard

PY - 2017

Y1 - 2017

N2 - We address two important issues in causal discovery from nonstationary or heterogeneous data, where parameters associated with a causal structure may change over time or across data sets. First, we investigate how to efficiently estimate the "driving force" of the nonstationarity of a causal mechanism. That is, given a causal mechanism that varies over time or across data sets and whose qualitative structure is known, we aim to extract from data a low-dimensional and interpretable representation of the main components of the changes. For this purpose we develop a novel kernel embedding of nonstationary conditional distributions that does not rely on sliding windows. Second, the embedding also leads to a measure of dependence between the changes of causal modules that can be used to determine the directions of many causal arrows. We demonstrate the power of our methods with experiments on both synthetic and real data.

AB - We address two important issues in causal discovery from nonstationary or heterogeneous data, where parameters associated with a causal structure may change over time or across data sets. First, we investigate how to efficiently estimate the "driving force" of the nonstationarity of a causal mechanism. That is, given a causal mechanism that varies over time or across data sets and whose qualitative structure is known, we aim to extract from data a low-dimensional and interpretable representation of the main components of the changes. For this purpose we develop a novel kernel embedding of nonstationary conditional distributions that does not rely on sliding windows. Second, the embedding also leads to a measure of dependence between the changes of causal modules that can be used to determine the directions of many causal arrows. We demonstrate the power of our methods with experiments on both synthetic and real data.

KW - Causal discovery

KW - Data distribution shift

KW - Kernel mean embedding

KW - Nonstationary driving force

UR - http://www.scopus.com/inward/record.url?scp=85043993746&partnerID=8YFLogxK

U2 - 10.1109/ICDM.2017.114

DO - 10.1109/ICDM.2017.114

M3 - Conference paper (refereed)

C2 - 31068766

T3 - Proceedings - IEEE International Conference on Data Mining, ICDM

SP - 913

EP - 918

BT - Proceedings : 17th IEEE International Conference on Data Mining, ICDM 2017

A2 - KARYPIS, George

A2 - ALU, Srinivas

A2 - RAGHAVAN, Vijay

A2 - WU, Xindong

A2 - MIELE, Lucio

PB - IEEE

T2 - IEEE International Conference on Data Mining

Y2 - 18 November 2017 through 21 November 2017

ER -

HUANG B, ZHANG K, ZHANG J, SANCHEZ-ROMERO R, GLYMOUR C, SCHÖKOPF B. Behind Distribution Shift: Mining Driving Forces of Changes and Causal Arrows. In KARYPIS G, ALU S, RAGHAVAN V, WU X, MIELE L, editors, Proceedings : 17th IEEE International Conference on Data Mining, ICDM 2017. IEEE. 2017. p. 913-918. (Proceedings - IEEE International Conference on Data Mining, ICDM). Epub 2017 Dec 18. doi: 10.1109/ICDM.2017.114

Behind Distribution Shift: Mining Driving Forces of Changes and Causal Arrows

Abstract

Publication series

Conference

Bibliographical note

Funding

Keywords

Access to Document

Other files and links

Fingerprint

Cite this