2023
Michael B Giles
MLMC techniques for discontinuous functions
arXiv:2301.02882
Vincent PH Goverse, Jad Hamdan and Jared Tanner
Optimal approximation complexity of high-dimensional functions with neural networks
Karl Welzel and Raphael A Hauser
Generalizing Quasi-Newton Updates to Higher-Order Derivative Tensors
arXiv:2301.11678
2022
Ben Walker, Felix Krones, Ivan Kiskin, Guy Parsons, Terry Lyons and Adam Mahdi
Dual Bayesian ResNet: A Deep Learning Approach to Heart Murmur Detection
Accepted to appear in Computing in Cardiology
Coralia Cartis and Zhen Shao
Random-subspace adaptive cubic regularisation method for nonconvex optimisationa>
Accpeted paper for NeurIPS 2022 workshop "Order up! The Benefits of Higher-Order Optimization in Machine Learning”
Coralia Cartis, Jaroslav Fowkes and Zhen Shao
Randomised subspace methods for non-convex optimization, with applications to nonlinear least-squares arXiv:2211.09873
Csaba Toth, Darrick Lee, Celia Hacker and Harald Oberhauser
Capturing Graphs with Hypo-Elliptic Diffusions
Accepted to appear in Advances in Neural Information Processing Systems (NeurIPS 2022)
Elena Gal, Shaun Singh, Aldo Pacchiano,
Ben Walker,
Terry Lyons and Jakob Foerster
Unbiased decisions reduce regret: adversarial optimism for the bank loan problem
KeCheng Chen, Elena Gal, Hong Yan and Haoliang Li
Domain Generalization with small data
Masaki Adachi, Satoshi Hayakawa, Martin Jørgensen, Harald Oberhauser and Michael A Osborne
Fast Bayesian Inference with Batch Bayesian Quadrature via Kernel Recombination
Accepted to appear in Advances in Neural Information Processing Systems (NeurIPS 2022)
Mohamed R Ibrahim and Terry Lyons
ImageSig: A signature transform for ultra-lightweight image recognition
IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) 2022
Oliver Sheridan-Methven and Michael Giles
Approximating inverse cumulative distribution functions to produce approximate random variables
arXiv:2012.09715
Patric Bonnier, Chong Liu and Harald Oberhauser
Adapted topologies and higher rank signatures
Accepted to appear in Annals of Applied Probability
Satoshi Hayakawa, Harald Oberhauser and Terry Lyons
Positively weighted kernel quadrature via subsampling
Accepted to appear in Advances in Neural Information Processing Systems (NeurIPS 2022)
Wenqi Zhu and Coralia Cartis
Quartic Polynomial Sub-problem Solutions in Tensor Methods for Nonconvex Optimization
Accpeted paper for NeurIPS 2022 workshop "Order up! The Benefits of Higher-Order Optimization in Machine Learning”
Satoshi Hayakawa, Harald Oberhauser and Terry Lyons
Hypercontractivity meets Random Convex Hulls: Analysis of Randomized Multivariate Cubatures
arXiv:2210.05787 (11 Oct 2022)