Teodora Pandeva

prof_pic.jpeg

Hi! I am a PhD student at the Amsterdam Machine Learning Lab (AMLab) within the AI4Science initiative. My research project aims to develop causal discovery algorithms for inferring gene regulatory networks. It is a result of a multi-disciplinary effort of Leendert Hamoen and Martijs Jonker from Swammerdam Institute for Life Sciences, Joris Mooij (Korteweg-De Vries Institute for Mathematics) and Patrick Forré (AMLab).

My research aims to shed some light on questions such as how to combine datasets from different environments with non-biological noise or how to find sparse causal graphs in a high-dimensional data setting. Thus, my research projects combine ideas from domain adaptation, causal discovery, and high-dimensional statistics.

selected publications

  1. Multi-View Independent Component Analysis with Shared and Individual Sources
    Teodora Pandeva, and Patrick Forré
    Proceedings of the 39th Conference on Uncertainty in Artificial Intelligence (UAI-23) 2023
    arXiv
  2. E-Valuating Classifier Two-Sample Tests
    Teodora Pandeva, Tim Bakker, Christian A. Naesseth, and 1 more author
    arXiv preprint arXiv:2210.13027 2022
    arXiv