Identifying Causal Mechanism Shifts Under Additive Models with Arbitrary Noise

Xia, Yewei; Cui, Xueliang; Zhang, Hao; Ren, Yixin; Xie, Feng; Guan, Jihong; Wang, Ruxin; Zhou, Shuigeng

doi:10.24963/ijcai.2025/524

Identifying Causal Mechanism Shifts Under Additive Models with Arbitrary Noise

Yewei Xia, Xueliang Cui, Hao Zhang, Yixin Ren, Feng Xie, Jihong Guan, Ruxin Wang, Shuigeng Zhou

Download BibTeX Cite

Proceedings of the Thirty-Fourth International Joint Conference on Artificial Intelligence. Main Track. Pages 4706-4714.

https://doi.org/10.24963/ijcai.2025/524

Keywords

Knowledge Representation and Reasoning: KRR: Causality
Knowledge Representation and Reasoning: KRR: Learning and reasoning

Abstract

In many real-world scenarios, the goal is to identify variables whose causal mechanisms change across related datasets. For example, detecting abnormal root nodes in manufacturing, and identifying key genes that influence cancer by analyzing differences in gene regulatory mechanisms between healthy individuals and cancer patients. This can be done by recovering the causal structure for each dataset independently and then comparing them to identify differences, but the performance is often suboptimal. Typically, existing methods directly identify causal mechanism shifts based on linear additive noise models (ANMs) or by imposing restrictive assumptions on the noise distribution. In this paper, we introduce CMSI, a novel and more general algorithm based on nonlinear ANMs that identifies variables with shifting causal mechanisms under arbitrary noise distributions. Evaluated on various synthetic datasets, CMSI consistently outperforms existing baselines in terms of F1 score. Additionally, we demonstrate CMSI's applicability on gene expression datasets of ovarian cancer patients at different disease stages.

Yewei Xia, Xueliang Cui, Hao Zhang, Yixin Ren, Feng Xie, Jihong Guan, Ruxin Wang, and Shuigeng Zhou. 2025. Identifying Causal Mechanism Shifts Under Additive Models with Arbitrary Noise. In Proceedings of the Thirty-Fourth International Joint Conference on Artificial Intelligence, IJCAI-25. IJCAI, 4706–4714. https://doi.org/10.24963/ijcai.2025/524

@inproceedings{ijcai2025p524,
  title     = {Identifying Causal Mechanism Shifts Under Additive Models with Arbitrary Noise},
  author    = {Xia, Yewei and Cui, Xueliang and Zhang, Hao and Ren, Yixin and Xie, Feng and Guan, Jihong and Wang, Ruxin and Zhou, Shuigeng},
  booktitle = {Proceedings of the Thirty-Fourth International Joint Conference on
               Artificial Intelligence, {IJCAI-25}},
  publisher = {International Joint Conferences on Artificial Intelligence Organization},
  editor    = {James Kwok},
  pages     = {4706--4714},
  year      = {2025},
  month     = {8},
  note      = {Main Track},
  doi       = {10.24963/ijcai.2025/524},
  url       = {https://doi.org/10.24963/ijcai.2025/524},
}

TY  - CPAPER
TI  - Identifying Causal Mechanism Shifts Under Additive Models with Arbitrary Noise
AU  - Xia, Yewei
AU  - Cui, Xueliang
AU  - Zhang, Hao
AU  - Ren, Yixin
AU  - Xie, Feng
AU  - Guan, Jihong
AU  - Wang, Ruxin
AU  - Zhou, Shuigeng
T2  - Proceedings of the Thirty-Fourth International Joint Conference on Artificial Intelligence, IJCAI-25
PB  - International Joint Conferences on Artificial Intelligence Organization
PY  - 2025
SP  - 4706
EP  - 4714
DO  - 10.24963/ijcai.2025/524
UR  - https://doi.org/10.24963/ijcai.2025/524
ER  -