← Authors

Laurence Aitchison

30 papers · 2014–2025 · 5 conferences · across top CS/AI conferences

Achievements

πŸƒ Academic Marathon (11) 🐣 Hot Topic Early Bird πŸŒ‰ Interdisciplinary Bridge πŸ—ΊοΈ Taxonomy Completionist (12) 🌈 Renaissance Researcher (8) 🧭 Keyword Pioneer 🐝 Cross-Pollinator (15) 🌍 Conference Polyglot (5) 🐺 Lone Wolf (4) πŸ‘‘ Triple Crown πŸ† Keyword Champion (3) πŸ‘₯ Mega-Team (25) πŸ”₯ Unstoppable (7) πŸ’Ž Century Club (30) ⚑ Prolific Year (5) πŸ—ƒοΈ Keyword Collector (73)

Conferences

ICML (9) ICLR (8) NIPS (8) UAI (4) CORL (1)

Top co-authors

Edward Milsom (5) Ben Anson (5) Xi Wang (3) Adam X. Yang (3) Vincent Fortuin (3) Sebastian W. Ober (3) AdriΓ  Garriga-Alonso (3) Maxime Robeyns (3) Conor Houghton (2) Mark van der Wilk (2)

Keywords

variational inference (7) bayesian inference (5) gaussian process (4) deep gaussian process (3) representation learning (3) bayesian neural network (3) deep wishart process (2) importance sampling (2) deep kernel machine (2) deep kernel process (2) latent variable model (2) variational autoencoder (2) kernel methods (2) supervised learning (1) text generation (1) approximate inference (1) data augmentation (1) wishart distribution (1) multi-task learning (1) image classification (1)

Papers

Residual Stream Analysis with Multi-Layer SAEs ICLR 2025
Position: Don’t Use the CLT in LLM Evals With Fewer Than a Few Hundred Datapoints ICML 2025
How to set AdamW’s weight decay as you scale model and dataset size ICML 2025
Function-Space Learning Rates ICML 2025
Jacobian Sparse Autoencoders: Sparsify Computations, Not Just Activations ICML 2025
Using Autodiff to Estimate Posterior Moments, Marginals and Samples UAI 2024
Instruction Tuning With Loss Over Instructions NIPS 2024
Stochastic Kernel Regularisation Improves Generalisation in Deep Kernel Machines NIPS 2024
Convolutional Deep Kernel Machines ICLR 2024
Bayesian Low-rank Adaptation for Large Language Models ICLR 2024
Position: Bayesian Deep Learning is Needed in the Age of Large-Scale AI ICML 2024
A theory of representation learning gives a deep generalisation of kernel methods ICML 2023
Semi-supervised learning with a principled likelihood from a generative model of data curation ICLR 2023
Robustness to corruption in pre-trained Bayesian neural networks ICLR 2023
Massively parallel reweighted wake-sleep UAI 2023
An improved variational approximate posterior for the deep Wishart process UAI 2023
Taylor TD-learning NIPS 2023
Bayesian Neural Network Priors Revisited ICLR 2022
Data augmentation in Bayesian neural networks and the cold posterior effect UAI 2022
Global inducing point variational posteriors for Bayesian neural networks and deep Gaussian processes ICML 2021
A variational approximate posterior for the deep Wishart process NIPS 2021
A statistical theory of cold posteriors in deep neural networks ICLR 2021
Tactile Image-to-Image Disentanglement of Contact Geometry from Motion-Induced Shear CORL 2021
Deep kernel processes ICML 2021
Why bigger is not always better: on finite and infinite neural networks ICML 2020
Bayesian filtering unifies adaptive and non-adaptive neural network optimization methods NIPS 2020
Tensor Monte Carlo: Particle Methods for the GPU era NIPS 2019
Deep Convolutional Networks as shallow Gaussian Processes ICLR 2019
Model-based Bayesian inference of neural activity and connectivity from all-optical interrogation of a neural circuit NIPS 2017
Fast Sampling-Based Inference in Balanced Neuronal Networks NIPS 2014