We are a research group in Data Science and Artificial Intelligence focusing on causal discovery.
Our goal is to establish causal discovery as a standard analytical methodology in scientific research, industry, and public policy, and to build a foundation in which causal inference becomes routinely used in society.
To achieve this goal, we collaborate with researchers from a wide range of disciplines. We work on problems arising in both fundamental sciences (such as natural and social sciences) and applied sciences (including engineering and medicine), contributing to their solutions from a methodological perspective.
A central focus of our research is the development of mathematical methodologies for uncovering causal mechanisms underlying natural phenomena and human behavior. In particular, we develop mathematical methods for inferring causal relationships from purely observational data without interventions, aiming to construct new methodological frameworks that overcome the limitations of conventional approaches.
因果探索を中心に研究するデータサイエンス・AIの研究室です。
本研究室では、因果探索を科学研究や産業・政策の現場における標準的な分析手法として定着させ、社会において因果推論が日常的に活用される基盤の構築を目指しています。
そのため、様々な科学分野の研究者と協力し、自然科学・社会科学などの基礎科学から工学・医学などの応用科学まで幅広い分野の問題に取り組み、方法論の立場から問題解決に貢献します。
特に、自然現象や人間行動の根底にある因果メカニズムを解明するための数理的方法論の研究を行っています。介入のない観察データから因果関係を推測するための数学的方法論を研究開発し、従来の限界を超える新しい方法論体系の構築を目指しています。
Professors | 教員
Administrative assistants | 事務補佐員
OKADA Hiroko | 岡田 拡子
Master's and Undergraduate Students | 博士前期課程学生・学部生
7 master's students | 博士前期課程学生 7名
4 undergraduates | 学部生 4名
Researchers
Visiting Researchers
Technical Assistant
Former staff | 旧スタッフ
Researchers | 研究者
SHIMIZU Shohei | 清水 昌平
(team leader | チームリーダー)
Professors | 教員
Ph.D. Students | 博士後期課程学生
SAKAMOTO Yuji | 坂本 雄司
TAGAWA Kentaro | 田川 健太郎
MORINISHI Yoshimitsu | 森西 美光
Master's Students | 博士前期課程学生
4 master's students | 博士前期課程学生 4名
Commissioned researchers | 受託研究員
Researchers | 研究者
SHIMIZU Shohei | 清水 昌平
(team director | チームディレクター)
Visiting Scientists | 客員研究員
Student trainee (RIKEN-AIP Overseas Student Collaboration Program)
KHAN Mariyam
Symposium: Approaching the World through the Lens of Causality: Slides
横浜国立大学統計的因果推論研究拠点 | YNU Research Center for Statistical Causal Inference
Financial literacy may not directly drive investment participation or retirement planning in Japan
Y. Jiang and S. Shimizu. Frontiers in Behavioral Economics, 2026.
科学技術・イノベーション政策の立案・評価への統計的因果探索の応用
高山・清水, 研究 技術 計画, 2025.
第58回IBISML研究会 企画セッション 「複雑データの基盤構造を探る—因果的抽出と次元の呪い—」 (2025年12月22日開催)
大規模言語モデルを活用した統計的因果探索の新展開 ― 統計的因果プロンプティングによる専門知識とデータの融合 ―.
三内・高山・清水. SBI金融経済研究所所報, 2025.
Differentiable causal discovery of linear non-Gaussian acyclic models under unmeasured confounding
Y. Morinishi and S. Shimizu, TMLR, 2025.
2025年いちょう祭での工学研究科に関する研究室見学時間は 5/2 10:00-13:00
Integrating large language models in causal discovery: a statistical causal approach
M. Takayama, T. Okuda, T. Pham, T. Ikenoue, S. Fukuma, S. Shimizu, A. Sannai, TMLR, 2025
Causal-discovery-based root-cause analysis and its application in time-series prediction error diagnosis
H. Yokoyama, R. Shingaki, K. Nishino, S. Shimizu, and T. Pham, Arxiv preprint, 2024. Accepted in Proc. IJCNN2025
Causal models and prediction in cell line perturbation experiments
J. Long, Y. Yang, S. Shimizu, T. Pham, and K.-A. Do, BMC Bioinformatics, 2025
ゼミ生の辻さん、武田さん、畑中さんがNEC Analytics Challenge Cup 2024 データ活用コンテストで優秀賞
Causal discovery with hidden variables based on non-Gaussianity and nonlinearity
T. N. Maeda, Y. Zeng, and S. Shimizu, Dependent Data in Social Sciences Research, 2024
Use of prior knowledge to discover causal additive models with unobserved variables and its application to time series data
T. N. Maeda and S. Shimizu, Behaviormetrika, 2024
6th Pacific Causal Inference Conference (PCIC 2024), Shanghai, China
Causal-learn: Causal discovery in Python
Y. Zheng, B. Huang, W. Chen, J. Ramsey, M. Gong, R. Cai, S. Shimizu, P. Spirtes, K. Zhang, JMLR Machine Learning Open Source Software, 2024
Counterfactual explanations of black-box machine learning models using causal discovery with applications to credit rating
D. Takahashi, S. Shimizu, and T. Tanaka, IJCNN2024. Accepted.
Scalable counterfactual distribution estimation in multivariate causal models
T. Pham, S. Shimizu, H. Hino, T. Le, CLeaR2024
Causal discovery with hidden variables based on non-Gaussianity and non-linearity
T. N. Maeda, Y. Zeng, and S. Shimizu. In Dependent Data in Social Sciences Research (2nd edition), Springer, 2024
Structure Learning for Groups of Variables in Nonlinear Time-Series Data with Location-Scale Noise
G. Kikuchi and S. Shimizu, Causal Analysis Workshop 2023 (CAWS2023)
Linkages among the Foreign Exchange, Stock, and Bond Markets in Japan and the United States
Y. Jiang and S. Shimizu, Causal Analysis Workshop 2023 (CAWS2023)
Causal Discovery for Non-stationary Non-linear Time Series Data Using Just-In-Time Modeling
D. Fujiwara, K. Koyama, K. Kiritoshi, T. Okawachi, T. Izumitani, S. Shimizu, CLeaR2023
Prospects of Continual Causality for Industrial Applications
D. Fujiwara, K. Koyama, K. Kiritoshi, T. Okawachi, T. Izumitani, S. Shimizu, First AAAI Bridge Program on Continual Causality, 2023
Differentiable causal discovery under heteroscedastic noise
G. Kikuchi, ICONIP2022
Python package for causal discovery based on LiNGAM
T. Ikeuchi, M. Ide, Y. Zeng, T. N. Maeda, S. Shimizu, JMLR Machine Learning Open Source Software, 2023
Statistical Causal Discovery: LiNGAM Approach
S. Shimizu, SpringerBriefs in Statistics, 2022
CNN-GRU based deep learning model for demand forecast in retail industry
K. Honjo, X. Zhou, S. Shimizu, IJCNN2022
Causal Discovery for Linear Mixed Data
Y. Zeng, S. Shimizu, H. Matsui, F. Sun, CLeaR2022
Python code
A Multivariate Causal Discovery based on Post-Nonlinear Model
K. Uemura, T. Takagi, T. Kambayashi, H. Yoshida, S. Shimizu, CLeaR2022
Python code
International Workshop on Causality and Philosophy, Online
4th March 2022 16:00-18:00 JST
15:00-17:00 in Hong Kong
8:00-10:00 in Berlin
統計的因果探索: 領域知識とデータから因果仮説を探索する
2021年度 JST-理研 合同AIP公開シンポジウム
International Symposium on Causal Inference and Machine Learning September 10-11, 2021
The KDD2021 Workshop on Causal Discovery (CD2021)
Video (Preliminary Version)
Estimating individual-level optimal causal interventions combining causal models and machine learning models
K. Kiritoshi, T. Izumitani, K. Koyama, T. Okawachi, K. Asahara, and S. Shimizu
Proc. 2021 ACM SIGKDD Workshop on Causal Discovery (CD2021)
Causal additive models with unobserved variables
T. N. Maeda and S. Shimizu, UAI2021
Python code
Causal discovery with multi-domain LiNGAM for latent factors
Y. Zeng, S. Shimizu, R. Cai, F. Xie, M. Yamamoto, and Z. Hao, IJCAI2021
Python code
[27th AIP Open Seminar] Talks by Causal Inference Team on 2nd June, 2021
The recording available
NeurIPS 2020 Workshop on Causal Discovery and Causality-Inspired Machine Learning
RCD: Repetitive causal discovery of linear non-Gaussian acyclic models with latent confounders
T. N. Maeda and S. Shimizu, AISTATS2020
Python code