Data management for machine learning (ML), optimization of ML pipelines, ML model management and materialization, learned and workload-aware indexing, graph processing.
To see what kind of research I’m currently interested in, see the articles below:
- “Certifiable Unlearning Pipelines for Logistic Regression: An Experimental Study“. Ananth Mahadevan, Michael Mathioudakis. DOI: https://doi.org/10.3390/make4030028.
- “Workload-Aware Materialization of Junction Trees“. Martino Ciaperoni, Cigdem Aslay, Aristides Gionis, Michael Mathioudakis. EDBT 2022. DOI: https://dx.doi.org/10.5441/002/edbt.2022.06
- “Workload-Aware Materialization for Efficient Variable Elimination on Bayesian Networks“. Cigdem Aslay, Martino Ciaperoni, Aristides Gionis, Michael Masthioudakis. ICDE 2021. DOI: 10.1109/ICDE51399.2021.00104
You want to request a review from me? Read this first.
Short Bio Before Helsinki, I taught at INSA Lyon for one semester (2017) and spent four years (2013-2017) as a postdoctoral researcher at Aalto University. I completed my doctoral studies at the University of Toronto (2013) and diploma of engineering at the National Technical University of Athens (2006). Awards: Best paper award at the ACM Web Conference 2022 and Springer Discovery Science 2020. Best student paper award at ACM WSDM 2016 and Web Science 2017.
NEW: Special Issue on “Data Processing for Machine Learning” in the Journal of Applied Sciences, MDPI.