Lennie Wells

Lennie Wells

he/him

StatML PhD student

Selected Publications

You can also find my articles on my Google Scholar profile.

Preprints

Regularised Canonical Correlation Analysis: graphical lasso, biplots and beyond

Published in arXiv preprint arXiv:2403.02979, 2024

This paper concerns how to regularise canonical correlation analysis to apply it in high dimensions. We propose a new method using the graphical lasso, and various visualisation techniques.

Recommended citation: L Wells, K Thurimella, S Bacallado. (2024). "Regularised Canonical Correlation Analysis: graphical lasso, biplots and beyond." arXiv preprint arXiv:2403.02979.
Download Paper

Conference Papers

Unconstrained Stochastic CCA: Unifying Multiview and Self-Supervised Learning

Published in The Twelfth International Conference on Learning Representations, 2023

We propose a novel formulation of Canonical Correlation Analysis (CCA) that leads to fast algorithms for Stochastic CCA and Deep CCA. We show this also immediately gives a new Self-Supervised Learning (SSL) method and explore its close relationship to an existing family of SSL methods.

Recommended citation: Chapman, J., Wells, L., & Lawry Aguila, A. (2024). "Unconstrained Stochastic CCA: Unifying Multiview and Self-Supervised Learning." The Twelfth International Conference on Learning Representations (ICLR).
Download Paper

Workshop Papers