Sotabase

Career

· PhD Student, MLG Cambridge2023–

Publications (21)

Laplace Redux - Effortless Bayesian Deep Learning
Neural Information Processing Systems · 2021
393
cited
Practical Deep Learning with Bayesian Principles
Neural Information Processing Systems · 2019
267
cited
Continual Deep Learning by Functional Regularisation of Memorable Past
Neural Information Processing Systems · 2020
164
cited
Benchmarking Neural Network Training Algorithms
arXiv.org · 2023
44
cited
Kronecker-Factored Approximate Curvature for Modern Neural Network Architectures
Neural Information Processing Systems · 2023
32
cited
Mixtures of Laplace Approximations for Improved Post-Hoc Uncertainty in Deep Learning
arXiv.org · 2021
28
cited
Approximate Bayesian Neural Operators: Uncertainty Quantification for Parametric PDEs
Trans. Mach. Learn. Res. · 2022
21
cited
Can We Remove the Square-Root in Adaptive Gradient Methods? A Second-Order Perspective
International Conference on Machine Learning · 2024
19
cited
Influence Functions for Scalable Data Attribution in Diffusion Models
International Conference on Learning Representations · 2024
19
cited
Accelerating Neural Network Training: An Analysis of the AlgoPerf Competition
International Conference on Learning Representations · 2025
18
cited
Posterior Refinement Improves Sample Efficiency in Bayesian Neural Networks
Neural Information Processing Systems · 2022
15
cited
Promises and Pitfalls of the Linearized Laplace in Bayesian Optimization
arXiv.org · 2023
12
cited
Purifying Shampoo: Investigating Shampoo's Heuristics by Decomposing its Preconditioner
arXiv.org · 2025
9
cited
Structured Inverse-Free Natural Gradient: Memory-Efficient & Numerically-Stable KFAC for Large Neural Nets
arXiv.org · 2023
7
cited
Position: Curvature Matrices Should Be Democratized via Linear Operators
arXiv.org · 2025
5
cited
Structured Inverse-Free Natural Gradient Descent: Memory-Efficient & Numerically-Stable KFAC
International Conference on Machine Learning · 2023
4
cited
Kronecker-factored Approximate Curvature (KFAC) From Scratch
arXiv.org · 2025
2
cited
Understanding and Improving Shampoo and SOAP via Kullback-Leibler Minimization
arXiv.org · 2025
2
cited
Better Hessians Matter: Studying the Impact of Curvature Approximations in Influence Functions
arXiv.org · 2025
In�uence Functions for Scalable Data Attribution in Di�usion Models
Sotabase
Runa Eschenhagen | Researcher Profile | Sotabase | Sotabase