Writing

Recent Writing

• June 11, 2026 Beta's memory length in weighted least squares regression

  You can use least squares regression to estimate the trend of a signal.If your inputs are time indices, your target is the signal itself, then the regression coefficie...

• April 29, 2021 Improving VAEs' Robustness to Adversarial Attacks – Part 2/2

  This is part two of a two-part series where I explain in some detail the ideas in my upcoming ICLR 2021 paper `Improving VAEs’ Robustness to Adversarial Attacks’, work...

• February 7, 2021 Learning Independent Representations using Flows & Fixed Linear Models

  This post gives an overview of the ideas in my AISTATS 2021 paper `Learning Bijective Feature Maps for Linear ICA’, work done with my friends and colleagues at the Uni...

• January 27, 2021 Improving VAEs' Robustness to Adversarial Attacks – Part 1/2

  This is part one of a two part series where I explain in some detail the ideas in my upcoming ICLR 2021 paper `Improving VAEs’ Robustness to Adversarial Attacks’, work...

• December 1, 2020 Explicit Regularisation in GNIs

  My 2020 NeurIPS paper, written with my friends and colleagues at the University of Oxford, studies the often used and yet poorly understood method of adding Gaussian n...