I am a data scientist passionate about monitoring and explainability of machine learning systems. In recent years, I have been working on developing tools and measures for issues surrounding fairness, privacy, and explainability in ML.

I am broadly interested in finding robust and trustworthy data-driven solutions to problems in healthcare, finance, social sciences, and systems biology. I have previously worked on developing novel computational approaches for recommender systems and for the analysis of genomics (Single-cell RNA) data.

I received a PhD in computer science from Boston University where my advisor was Professor Mark Crovella. During my PhD, I was also a research visitor at the Max Planck Institute for Software Systems where I worked with Professor Krishna P. Gummadi. Before that, I received a M.Sc. in computer science from University of Bonn and a B.Sc. in software engineering from University of Isfahan.

Updates and News

• November 2021. I joined Fiddler AI.
• September 2021. Our paper “Auditing Black-Box Prediction Models for Data Minimization Compliance” has been accepted for a spotlight presentation (<3% acceptance rate) at NeurIPS2021.
• August 2021. I successfully defended my PhD dissertation, “Tools for Responsible Decision Making in Machine Learning”.
• Juy 2020. I presented our paper about the incompatibility of fairness in privacy and accuracy at the FairUMAP workshop in ACM UMAP 2020.
• April 2020. Our novel network-guided decomposition technique for the analysis of scRNA data is published in Nature Communications. The paper is available here.
• February 2019. I presented our paper about using antidote data to improve social properties of recommender systems at the ACM International Conference on Web Search and Data Mining (WSDM’19).