Teodora Reu

DPhil in Computer Science at University of Oxford

prof_pic.jpeg

Hey! I’m Teo, currently in my first year of DPhil in Computer Science, working under the guidance of Professor Michael Bronstein. My research focuses on Geometric Deep Learning, Diffusion Models, and spectral Graph Neural Networks (GNNs).

Before this, I did my MPhil in Advanced Computer Science at the University of Cambridge, where I graduated with distinction. During my time there, I dug into the world of Graph Neural Networks, specifically looking into variational and attentional approaches to graph rewiring. I was lucky to work with Professor Pietro Liò, Francesco Di Giovanni, and Chaitanya Joshi. I also dabbled in theoretical aspects of diffusion samplers under the guidance of Francisco Vargas. Looking forward to the exciting journey ahead!

My industry experience is varied from big tech, to start ups, and then to finance company, I worked as a software engineer summer intern at Google, R3, and BlackRock.

PS: I am also a very exited and passionate mathematician! I love mathematics! Was very fortunate to work as a counsellor this summer at PROMYS, helping international olympiad level highschoolers discover Number Theory and Group Theory from first principles.

selected publications and projects

  1. diff.png
    Expressiveness Remarks for Denoising Diffusion Based Sampling
    Francisco Vargas, Teodora Reu, and Anna Kerekes
    In Fifth Symposium on Advances in Approximate Bayesian Inference, 2023
  2. cin.png
    CIN++: Enhancing Topological Message Passing
    Lorenzo Giusti, Teodora Reu, Francesco Ceccarelli, Cristian Bodnar, and 1 more author
    arXiv preprint arXiv:2306.03561, 2023
  3. var_atte.png
    Rethinking Graph Topology Attentional and Variational Approaches
    Teodora Reu, and Pietro Liò
    2023
  4. breast_cancer.png
    Graph Neural Networks for Breast Cancer Data Integration
    Teodora Reu
    arXiv preprint arXiv:2211.15561, 2022
  5. fisher.png
    An Exploration of Lottery Ticket Hypopthesis thorough Fisher Prunning
    Teodora Reu, Thomas Christie, Yumna Naqvi, and Matevz Matjasec
    Theory of Deep Learning Lecture, 2023
  6. stable_star_systems.png
    Looking for Stable Celestial Systems Using Bayesian Optimisation
    Eirik Fladmark, Teodora Reu, and Laura Brinkholm Justesen
    arXiv preprint arXiv:2303.14835, 2023