Osman Ali Mian
CISPA Helmholtz Center For Information Security
Email: osmanalimian [at] outlook [dot] com
About
My research is about causal discovery and inference, and to reason how it can be used to make machine learning robust. I work on inventing privacy-preserving, federated causal discovery algorithms under mild assumptions such that these methods can actually be applied to real-world scenarios. You can view my CV, last updated in March 2024, here.
News
[17-May-2024] Our work on causal discovery from continually arriving episodic data was accepted at ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (KDD), ACM, 2024.
[27-Mar-2024] I will give a talk at the Trustworthy Machine Learning group at IKIM – Institute for Artificial Intelligence in Medicine in Essen on 15-Apr-2024.
[22-Mar-2024] I will give a talk at CAUSE Junior Researcher Day Spring 2024 at Technical University of Munich (TUM) on 11-Apr-2024.
[03-Jan-2024] Our work on causal discovery from episodic data was accepted at The Fourth Workshop of Artificial Intelligence for Time Series Analysis (AI4TS) at AAAI 2024.
[20-Dec-2023] Our work on continual learning of causal networks was accepted for oral presentation at The Second AAAI Bridge Program on Continual Causality at AAAI 2024.
[25-Apr-2023] I will be attending AISTATS 2023 in Valencia, Spain to present, PERI, our proposed approach for Federated Causal Discovery.
[22-Feb-2023] I gave a talk at the Causality Discussion Group where I presented ORION, Our proposed approach for causal discovery and intervention detection over multiple environments. You can find the recorded talk here.
[21-Nov-2022] I will be working as an Applied Science Intern at Amazon from Dec 2022 until April 2023. The internship will be about applying Causal Learning to Natural Language Understanding.
[02-Sep-2022] I had a chance to talk about my work on causal machine learning in TL;DR, CISPA's first podcast series. You can stream the podcast here.