Charlie B. Tan

DPhil Student @ University of Oxford

prof_pic.png

Hi there! I’m Charlie, a second-year DPhil Computer Science student at the University of Oxford. I’m interested in the overlap between generative modeling and unnormalized density sampling, particularly for scientific applications. My research is supervised by Prof Michael Bronstein, and funded by a Departmental Studentship.

Last summer I interned at InstaDeep, working on generative protein design in the Bayesian Flow Networks team.

Prior to Oxford, I studied the MPhil in Advanced Computer Science at the University of Cambridge, graduating with distinction. Whilst in Cambridge my research focused on deep learning theory, and was supervised by Prof. Ferenc Huszár. I was particularly interested in geometric and topological theories of generalisation. My MPhil dissertation concerned approximate second-order optimisation, and the bias of such optimisers towards generalisation.

[name](dot)[surname]@cs.ox.ac.uk

news

Nov 3, 2024 New preprint - applying techniques developed for my MPhil dissertation to RL!
Nov 1, 2024 Camera ready version of On the Limitations of Fractal Dimension as a Measure of Generalization released
Sep 26, 2024 On the Limitations of Fractal Dimension as a Measure of Generalization accepted to NeurIPS 2024!
May 23, 2024 Started a research internship at InstaDeep, in the Bayesian Flow Networks team.
Oct 2, 2023 I have joined Prof. Michael Bronstein’s group for my DPhil at the University of Oxford

selected projects

  1. fractal.png
    On the Limitations of Fractal Dimension as a Measure of Generalization
    Charlie B. Tan, Inés García-Redondo, Qiquan Wang, Michael M. Bronstein, and Anthea Monod
    2024
  2. ppo.png
    Beyond the Boundaries of Proximal Policy Optimization
    Charlie B. Tan, Edan Toledo, Benjamin Ellis, Jakob N. Foerster, and Ferenc Huszár
    2024
  3. geodesic.jpg
    Geodesic Mode Connectivity
    Charlie B. Tan, Theodore Long, Sarah Zhao, and Rudolf Laine
    ICLR TinyPapers, 2023