Algorithms, Applications, and Verification of Causal Structure Learning

CSE Best PhD Dissertation Award Seminar

Speaker: Dr. Pingchuan MA
Professor, Zhejiang University of Technology
and
CSE Best PhD Dissertation Award (2024-25) Recipient
The Hong Kong University of Science and Technology

Title: Algorithms, Applications, and Verification of Causal Structure Learning

Date: Monday, 20 April 2026

Time: 4:30pm - 5:30pm

Venue: Lecture Theater F
(Leung Yat Sing Lecture Theater), near lift 25/26, HKUST

Abstract:

Randomized experiments are the gold standard for uncovering causal relationships, but they are often costly, impractical, or ethically infeasible. This talk presents a data-driven and verifiable approach to causal structure learning from observational data, with a focus on improving accuracy, interpretability, and reliability.

The talk first introduces SPOT, a scalable causal discovery method that infers posterior distributions over causal skeletons and uses them to guide graph search in a continuous optimization framework, achieving strong performance in both accuracy and scalability. It then presents XINSIGHT, an end-to-end framework for explainable data analysis that leverages causal structure learning to generate qualitative and quantitative explanations for data facts. Finally, the talk introduces CICHECK, a runtime verification framework that strengthens causal discovery algorithms by detecting erroneous conditional independence tests and pruning excessive ones, thereby improving both reliability and privacy.

Overall, the talk highlights how causal discovery can be advanced not only through better learning algorithms, but also through integration with explainable data systems and formal verification techniques, moving the field closer to practical and trustworthy real-world deployment.

Biography:

Dr. Pingchuan MA is a Professor in the College of Computer Science and Technology at Zhejiang University of Technology. His research lies at the intersection of data systems, causal inference, and privacy-enhancing computation. He is also the co-founder of CipherInsight Limited, a HKUST spin-off focusing on verifiable and privacy-preserving analytics. His research has been recognized by multiple international venues such as SIGMOD, ICSE, KDD, and VLDB, and he has received honors including the HKUST SENG PhD Research Excellence Award and the Hong Kong ICT Bronze Award.