Deep kernel methods
Schedule
Thursday, November 25, 2021
13:00 UTC
Laurence Aitchison
Dr
University of Bristol
Host: Sheffield ML
Recording
Open recording
Event Information
Recording
Available
Host
Sheffield ML
Duration
70 minutes
Abstract
Deep neural networks (DNNs) with the flexibility to learn good top-layer representations have eclipsed shallow kernel methods without that flexibility. Here, we take inspiration from deep neural networks to develop a new family of deep kernel method. In a deep kernel method, there is a kernel at every layer, and the kernels are jointly optimized to improve performance (with strong regularisation). We establish the representational power of deep kernel methods, by showing that they perform exact inference in an infinitely wide Bayesian neural network or deep Gaussian process. Next, we conjecture that the deep kernel machine objective is unimodal, and give a proof of unimodality for linear kernels. Finally, we exploit the simplicity of the deep kernel machine loss to develop a new family of optimizers, based on a matrix equation from control theory, that converges in around 10 steps.
Topics
Related Job Opportunities
JobJobJob
PhD Studentship: Mitochondrial Metabolism and Novel Therapeutic Strategies for Metabolic Dysfunction-Associated Steatotic Liver Disease (MASLD) (Fixed Term)
A fully funded University of Cambridge PhD studentship, supported by Novo Nordisk, will investigate how mitochondrial metabolism changes during metabolic dysfunction-associated steatotic liver…
Posted Aug 10, 2026
Research Associate (Fixed Term)
Kathy Niakan's laboratory at the Loke Centre for Trophoblast Research is recruiting a postdoctoral researcher to study early lineage specification in human pre- and early post-implantation embryos.…
Posted Aug 11, 2026
Cambridge, United Kingdom
Research Assistant/Associate (Fixed Term)
A fixed-term research position in the laboratories of Ole Paulsen and Jasper Poort will study neural mechanisms of visual learning in mice. The project combines patch-clamp electrophysiology…
Posted Aug 11, 2026
Cambridge, United Kingdom