About

I am currently a Research Scientist in the Adaptive Experimentation team at Meta, where we develop sample-efficient methods for block-box optimization problems. My research focuses on hyperparameter tuning and transfer for deep learning, including learning curve prediction and neural scaling laws. Previously, I completed my PhD in the Computational and Biological Learning Lab at the University of Cambridge, where I worked on probabilistic machine learning under the supervision of José Miguel Hernández-Lobato. During my PhD, I was funded by the Harding Distinguished Postgraduate Scholars Programme and affiliated with the Empirical Inference department at the Max Planck Institute for Intelligent Systems. Before starting my PhD, I received my B.Sc. in Computer Science and M.Sc. in Autonomous Systems from TU Darmstadt in Germany, where I worked with Stefan Roth, Jan Peters and Kristian Kersting. I also attended The University of British Columbia, Stanford University and Aalto University as a visiting student, and participated in Fulbright’s Leaders in Entrepreneurship program.