News
The full-length version of our ICML workshop paper, Parameter symmetries determine representational geometry in overparameterized nonlinear networks, has been accepted at NeurIPS 2026 and is now available on arXiv.
↗
I attended the Visual Neuroscience: From Spikes to Awareness summer school at Schloss Rauischholzhausen from 30 August to 11 September 2026. The programme featured lectures by leading vision researchers, including Edward Adelson, Tony Movshon, and Kalanit Grill-Spector.
At CCN 2026 in New York, I co-organised the keynote & tutorial session Adaptive, degenerate, and yet comparable? Rethinking representational comparisons with Erin Grant, Lukas Braun, and Eleanor Holton. I also presented two posters: one on how parameter symmetries reshape representational geometry in nonlinear networks while the realised function remains unchanged, and another on the dissociation between function and representation in task-optimised vision networks.
↗
We received the Best Paper Award for our work, Parameter symmetries determine representational geometry in overparameterized nonlinear networks, at the ICML 2026 Workshop on Weight-Space Symmetries: From Foundations to Practical Applications that took place in Seoul, South Korea, on Friday 10 July 2026.
↗
Our work, Parameter symmetries determine representational geometry in overparameterized nonlinear networks, was accepted as an oral presentation at the ICML 2026 Workshop on Weight-Space Symmetries: From Foundations to Practical Applications taking place in Seoul, South Korea, on Friday 10 July 2026.
↗
Our proposal, Adaptive, degenerate, and yet comparable? Rethinking representational comparisons, led by Erin Grant and Lukas Braun, has been accepted as one of three keynote & tutorial sessions at the 9th Annual Conference on Cognitive Computational Neuroscience. The session will take place in New York on Monday 3 August 2026.
↗
I have joined the Neural Information Processing Group at the University of Tübingen and the International Max Planck Research School for Intelligent Systems (IMPRS-IS) as a doctoral researcher.
↗
I am excited to be starting a position as a project research intern at the Gatsby Computational Neuroscience Unit at UCL. Together with Erin Grant, I will work on representational geometry in the Theory of Learning Lab, led by Andrew M. Saxe.