Sascha Xu

PhD Student
Exploratory Data Analysis
CISPA Helmholtz Center for Information Security
CISPA D2, Room 0.02
Im Oberem Werk 1
66386 St. Ingbert, Germany
sascha.xu@cispa.de

Sascha Xiaguang Xu is a Ph.D. student who works on the intercept between causality and explainability.

Sascha obtained his Bachelor's and Master's degrees from Saarland University, respectively in 2019 and 2022. During this time he worked with us a student research assistant on the topic of bivariate causal inference in the presence of heteroscedastic noise, which led to an ICML paper in 2022.

Publications

2026
Wilms, M, Xu, S & Vreeken, J Explainable Mixture Models through Differentiable Rule Learning. In: Proceedings of the International Conference on Representation Learning (ICLR), OpenReview, 2026. (28.2% acceptance rate)
Al Rahwanji, MJ, Xu, S, Walter, NP & Vreeken, J Learning and Naming Subgroups with Exceptional Survival Characteristics. Technical Report 2602.22179, arXiv, 2026.project website

2025
Xu, S, Cueppers, J & Vreeken, J Succinct Interaction-Aware Explanations. In: Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (KDD), pp 1715-1726, ACM, 2025. (19% acceptance rate)project website
Xu, S, Mameche, S & Vreeken, J Information-Theoretic Causal Discovery in Topological Order. In: Proceedings of the 28th International Conference on Artificial Intelligence and Statistics (AISTATS), pp 2008-2016, PMLR, 2025. (31.3% acceptance rate)project website
Xu, S, Walter, NP & Vreeken, J Neural Rule Lists: Learning Discretizations, Rules, and Order in One Go. In: Proceedings of Neural Information Processing Systems (NeurIPS), PMRL, 2025. (24.5% acceptance rate)project website

2024
Cueppers, J, Xu, S, Musa, A & Vreeken, J Causal Discovery from Event Sequences by Local Cause-Effect Attribution. In: Proceedings of Neural Information Processing Systems (NeurIPS), PMRL, 2024. (25.8% acceptance rate)project website
Xu, S, Walter, NP, Kalofolias, J & Vreeken, J Learning Exceptional Subgroups by End-to-End Maximizing KL-divergence. In: Proceedings of the International Conference on Machine Learning (ICML), PMLR, 2024. (spotlight, 3.5% acceptance rate; 27.5% overall)project website
Xu, S, Cueppers, J & Vreeken, J Succinct Interaction-Aware Explanations. Technical Report 2402.05566, arXiv, 2024.project website

2022
Xu, S, Mian, O, Marx, A & Vreeken, J Inferring Cause and Effect in the Presence of Heteroscedastic Noise. In: Proceedings of the International Conference on Machine Learning (ICML), PMLR, 2022. (21.9% acceptance rate)project website
Xu, S, Marx, A, Mian, O & Vreeken, J Causal Inference with Heteroscedastic Noise Models. In: Proceedings of the AAAI Workshop on Information Theoretic Causal Inference and Discovery (ITCI'22), 2022.