CMU-CLeaR Group

Welcome to the CMU-CLeaR Group! CLeaR (Causal Learning and Reasoning) is a research group led by Professors Kun Zhang, Peter Spirtes, Clark Glymour, and Joseph Ramsey at Carnegie Mellon University.

Research synopsis: Our research interests lie in machine learning and artificial intelligence, especially in causal discovery, causal representation learning, and causality-related ML and AI. On the causal learning side, we develop methods for automated causal discovery or representation learning from various kinds of data. Our research has delivered state-of-the-art causal discovery methodological developments by considering many of the challenges to causal learning, including latent confounding, distribution shifts, nonlinear relationships, general data distributions, and selection bias. On the ML/AI side, we investigate learning problems including transfer learning, concept learning, reinforcement learning, and deep learning from a causal view. On the philosophy side, we study philosophical foundations of causation and various machine learning tasks, and explore ethical issues of artificial intelligence. On the application side, we develop domain-specific causality algorithms to solve exciting real-world applications in neuroscience, biology, healthcare, computer vision, computational finance, and climate analysis.


04/2024 Ignavier Ng’s work on continuous optimization-based causal discovery received the Best Paper Award at CLeaR 2024. Yujia Zheng’s earlier work on domain adaptation received the Best Paper Award (Honorable Mention) at WSDM 2024. Congratulations to them!
01/2024 Haoyue Dai, Yujia Zheng, and Ignavier Ng were selected for oral presentations at ICLR 2024, NeurIPS 2023, and CLeaR 2024, respectively. The topics are gene regulatory network inference, nonlinear ICA, and continuous optimization-based causal discovery. Congratulations!
12/2023 Ever wonder your personality and its connections to other things? Contribute to our ongoing research on personality, physical features, and demographics by completing our survey (results are available at the end)! We appreciate your time, your curiosity, and your contribution!
05/2023 Kun Zhang and Peter Spirtes were lecturers for CBMS Conference – Foundations of Causal Graphical Models and Structure Discovery (with 10 lectures): slides and video recordings are available.
05/2023 We are excited to be part of AI Institute for Societal Decision Making (AI-SDM)! See CMU news and Dietrich news. We are looking forward to developing suitable methods for causal learning and counterfactual reasoning for transparent, trustworthy, and effective decision making.
04/2023 Peter Spirtes received the AWS AI Amazon Research Award for the 2022 cycle!
12/2022 Our team (Haoyue*, Ignavier*, Xinshuai*, Yujia, Biwei, Kun) ranked the 1st in the NeurIPS 2022 CausalML Challenge: Causal Insights for Learning Paths in Education competition! Specifically, we ranked the 1st, 1st, 1st, and 2nd, in the four tasks, respectively!
11/2022 Zeyu Tang has been selected as a K&L Gates Presidential Fellow in Ethics and Computational Technologies for the 2023-2024 Academic Year! Congratulations, Zeyu!
10/2022 Zeyu Tang and Yujia Zheng were chosen as Top Reviewers of NeurIPS 2022! Thanks for their service to the community!
08/2022 Many of our group members made excellent contributions to UAI 2022 as organizing committee members!
07/2022 Congratulations on Biwei’s successful PhD defense! She will join UCSD as an assistant professor in the fall.
07/2022 Zeyu Tang and Kun Zhang received Best Paper Award of the ICML 2022 Workshop on New Frontiers in Adversarial Machine Learning (AdvML-Frontiers@ICML 2022), together with Yatong Chen and Yang Liu.
07/2022 Kun Zhang received ICML 2022 Test of Time Award Honorable Mention (together with Bernhard Schölkopf, Dominik Janzing, Jonas Peters, Eleni Sgouritsa, Joris Mooij).
11/2021 We are excited to release causal-learn, a Python package for causal discovery!