1. 'Amortized Sampling with Transferable Normalizing Flows' accepted to NeurIPS 2025; 2. Presented 'Scalable Equilibrium Sampling with Sequential Boltzmann Generators' at ICML 2025; 3. 'On the Limitations of Fractal Dimension as a Measure of Generalization' accepted to NeurIPS 2024.
Research Experience
1. Currently visiting Mila - Quebec AI Institute, hosted by Dr. Kirill Neklyudov; 2. Interned at InstaDeep, working on generative protein sequence modeling.
Education
1. DPhil student at the University of Oxford, supervised by Prof. Michael Bronstein and Dr. Alex Tong; 2. MPhil in Advanced Computer Science from the University of Cambridge, graduated with distinction, research focused on deep learning theory, supervised by Prof. Ferenc Huszár.
Background
Research interests: generative methods for drawing samples from Boltzmann densities. Field: Computer Science.