Kiarash Mohammadi
Hi there! My name is Kiarash, a recent MSc graduate from Mila and the University of Montreal. I was supervised by Dr. Golnoosh Farnadi.
My main interest lies within the broad area of trustworthy AI, mainly fairness and explainability — problems that should have been thought about prior to the large deployment of AI/ML! I like to approach these problems from a causal perspective with an eye on the rich asset of theoretical CS.
I also like to cook, (struggle to) make Latte arts, go cycling and bouldering, listen to Persian classical, and explore gorgeous Montreal!
Education
Mila, Université de Montréal, MSc in
Computer Science - AI (2021 - 2023)Ferdowsi University of Mashhad, BSc in Computer Engineering (2016 - 2021)
Experience
DRW
Data Scientist Intern (Present)Oracle
ML Research Intern (Spring and Summer 2023)Borealis AI, RBC
ML Research Intern (Summer 2022)Max Planck Institute for Intelligent Systems
BSc Intern (Winter and Summer 2020)Johannes Kepler University Linz
BSc Intern (Summer 2019)IST Austria - BSc Intern (2018 - 2019)
Publications
"Ranking Regularization for Critical Rare Classes: Minimizing False Positives at a High True Positive Rate", K. Mohammadi, H. Zhao, M. Zhai, and F. Tung — Computer Vision and Pattern Recognition Conference (CVPR), 2023 (paper).
"Causal Adversarial Perturbations for Individual Fairness and Robustness in Heterogeneous Data Spaces", A. Ehyaei, K. Mohammadi, A. Karimi, S. Samadi, G. Farnadi — AAAI, 2024 (paper).
"FETA: Fairness Enforced Verifying, Training, and Predicting Algorithms for Neural Networks", K. Mohammadi, A. Sivaraman, and G. Farnadi — ACM Conference on Equity and Access in Algorithms, Mechanisms, and Optimization (EAAMO), 2023 (paper).
"Post-processing Counterexample-guided Fairness Guarantees in Neural Networks", K. Mohammadi, A. Sivaraman, and G. Farnadi — AAAI CLeaR Workshop, 2022.
"Scaling Guarantees for Nearest Counterfactual Explanations", K. Mohammadi, A. Karimi, G. Barthe, and I. Valera — AAAI/ACM Conference on AI, Ethics, and Society (AIES), 2021. (paper, oral, poster, code)
"Faster Algorithms for Quantitative Analysis of MCs and MDPs with Small Treewidth", A. Asadi, K. Chatterjee, A.K. Goharshady, K. Mohammadi, and A. Pavlogiannis — International Symposium on Automated Technology for Verification and Analysis (ATVA), 2020. (paper, oral, code)
Contact
Please feel free to reach out to me via kiarash {dot} km {at} gmail