Isaac Reid

I’m a third-year PhD student at the Machine Learning Group within the Cambridge University Engineering Department. I am interested in problems at the interface of ML, statistical physics and applied mathematics.

Donostia, April 2024

Donostia, April 2024

I’m co-supervised by Dr Adrian Weller and Prof. Rich Turner, and collaborate closely with Prof. Krzysztof Choromanski (Google DeepMind and Columbia). I’m funded by a Trinity College external studentship and a Google PhD Fellowship.

Research. I’m currently interested in efficient Transformer attention for graph-structured data. I’ve designed novel Monte Carlo estimators to approximate the softmax kernel, random walk-based algorithms for building ‘graph random features’, and new position encodings using Lie groups. I’m also a dilettante in robotics, statistical physics, approximate Bayesian inference, and influence functions.

Teaching. I teach mathematics for second-year undergraduate engineers at Trinity, and mentor fantastic master’s students. I was a guest lecturer for the 2024-2025 Cambridge MLMI course, providing an introduction to Transformers.

Past lives. I obtained my MPhys at Hertford College, Oxford, where I submitted my dissertation on entanglement barriers in dual-unitary quantum circuits, supervised by Dr Bruno Bertini and Prof. Fabian Essler. I’ve also researched simplicity bias in deep neural networks with Prof. Ard Louis at the Rudolf Peierls Centre for Theoretical Physics, Oxford, and Bose-Einstein condensation in active matter with Dr Benoît Mahault and Prof. Ramin Golestanian at the Max Planck Institute for Dynamics and Self Organisation, Göttingen. I worked at this early-stage startup for a year, developing adaptive optics for laser fabrication of qubits inside diamonds, and currently work part-time as a research associate at IQ Capital.

Contact me at ir337(at)cam.ac.uk.

News: