Downloads: 96

Files in This Item:
File Description SizeFormat 
djohk00815.pdfDissertation_全文2.92 MBAdobe PDFView/Open
yjohk00815.pdfAbstract_要旨166.12 kBAdobe PDFView/Open
Title: Treatment Effect Estimation from Small Observational Data
Other Titles: 小規模観察データからの介入効果推定
Authors: Harada, Shonosuke
Author's alias: 原田, 将之介
Keywords: Treatment Effect Estimation
Causal Inference
Machine Learning
Issue Date: 23-Mar-2023
Publisher: Kyoto University
Conferring University: 京都大学
Degree Level: 新制・課程博士
Degree Discipline: 博士(情報学)
Degree Report no.: 甲第24727号
Degree no.: 情博第815号
Conferral date: 2023-03-23
Degree Call no.: 新制||情||137(附属図書館)
Degree Affiliation: 京都大学大学院情報学研究科知能情報学専攻
Examination Committee members: (主査)教授 鹿島 久嗣, 教授 阿久津 達也, 教授 下平 英寿
Provisions of the Ruling of Degree: 学位規則第4条第1項該当
Rights: 3章は1及び4に基づく。 4章は2及び5に基づく。 5章は3に基づく。 1. Shonosuke Harada and Hisashi Kashima. “Counterfactual Propagation for Semi-Supervised Individual Treatment Effect Estimation”. In: Joint European Conference on Machine Learning and Knowledge Discovery in Databases. Springer. 2021, pp. 542–558. https://doi.org/10.1007/978-3-030-67658-2_31 2. Shonosuke Harada and Hisashi Kashima. “GraphITE: Estimating Individual Effects of Graph-Structured Treatments”. In: Proceedings of the 30th ACM International Conference on Information and Knowledge Management. 2021, pp. 659–668. https://doi.org/10.1145/3459637.3482349 3. Shonosuke Harada and Hisashi Kashima. “InfoCEVAE: Treatment Effect Estimation with Hidden Confounding Variables Matching”. In: Machine Learning (2022), pp. 1–19. https://doi.org/10.1007/s10994-022-06246-0 4. 原田将之介, 鹿島久嗣. ""反事実伝播: 介入効果推定のための半教師付き学習"". 人工知能学会論文誌 2022 年 37 巻 3 号 p. B-LA3_1-14. https://doi.org/10.1527/tjsai.37-3_B-LA3 5. 原田 将之介, 鹿島 久嗣. ""GraphITE: グラフ介入に対する介入効果推定"". 人工知能学会論文誌 2022 年 37 巻 6 号 p. D-M73_1-11. https://doi.org/10.1527/tjsai.37-2_D-M73
DOI: 10.14989/doctor.k24727
URI: http://hdl.handle.net/2433/283846
Appears in Collections:140 Doctoral Dissertation (Informatics)

Show full item record

Export to RefWorks


Export Format: 


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.