1 code implementation • 2 Mar 2020 • Mark van der Wilk, Vincent Dutordoir, ST John, Artem Artemev, Vincent Adam, James Hensman
One obstacle to the use of Gaussian processes (GPs) in large-scale problems, and as a component in deep learning system, is the need for bespoke derivations and implementations for small variations in the model or inference.
2 code implementations • 16 Feb 2023 • Victor Picheny, Joel Berkeley, Henry B. Moss, Hrvoje Stojic, Uri Granta, Sebastian W. Ober, Artem Artemev, Khurram Ghani, Alexander Goodall, Andrei Paleyes, Sattar Vakili, Sergio Pascual-Diaz, Stratis Markou, Jixiang Qing, Nasrulloh R. B. S Loka, Ivo Couckuyt
We present Trieste, an open-source Python package for Bayesian optimization and active learning benefiting from the scalability and efficiency of TensorFlow.
1 code implementation • 12 Apr 2021 • Vincent Dutordoir, Hugh Salimbeni, Eric Hambro, John McLeod, Felix Leibfried, Artem Artemev, Mark van der Wilk, James Hensman, Marc P. Deisenroth, ST John
GPflux is compatible with and built on top of the Keras deep learning eco-system.
1 code implementation • 14 Oct 2022 • Alexander Terenin, David R. Burt, Artem Artemev, Seth Flaxman, Mark van der Wilk, Carl Edward Rasmussen, Hong Ge
For low-dimensional tasks such as geospatial modeling, we propose an automated method for computing inducing points satisfying these conditions.
1 code implementation • 28 Jun 2022 • Artem Artemev, Tilman Roeder, Mark van der Wilk
We believe that further focus on removing memory constraints at a compiler level will widen the range of machine learning methods that can be developed in the future.
no code implementations • 15 Feb 2019 • Vincent Dutordoir, Mark van der Wilk, Artem Artemev, James Hensman
We also demonstrate that our fully Bayesian approach improves on dropout-based Bayesian deep learning methods in terms of uncertainty and marginal likelihood estimates.
no code implementations • 5 Dec 2019 • Victor Picheny, Sattar Vakili, Artem Artemev
Bayesian optimisation is a powerful tool to solve expensive black-box problems, but fails when the stationary assumption made on the objective function is strongly violated, which is the case in particular for ill-conditioned or discontinuous objectives.
no code implementations • 15 Jan 2020 • Vincent Adam, Stefanos Eleftheriadis, Nicolas Durrande, Artem Artemev, James Hensman
The use of Gaussian process models is typically limited to datasets with a few tens of thousands of observations due to their complexity and memory footprint.
no code implementations • NeurIPS 2021 • Sattar Vakili, Henry Moss, Artem Artemev, Vincent Dutordoir, Victor Picheny
We provide theoretical guarantees and show that the drastic reduction in computational complexity of scalable TS can be enjoyed without loss in the regret performance over the standard TS.
no code implementations • 25 Jun 2020 • Victor Picheny, Vincent Dutordoir, Artem Artemev, Nicolas Durrande
Many machine learning models require a training procedure based on running stochastic gradient descent.
no code implementations • 16 Feb 2021 • Artem Artemev, David R. Burt, Mark van der Wilk
We propose a lower bound on the log marginal likelihood of Gaussian process regression models that can be computed without matrix factorisation of the full kernel matrix.
no code implementations • NeurIPS Workshop ICBINB 2021 • David R. Burt, Artem Artemev, Mark van der Wilk
We suggest a method that adaptively selects the amount of computation to use when estimating the log marginal likelihood so that the bias of the objective function is guaranteed to be small.
no code implementations • pproximateinference AABI Symposium 2019 • Mark van der Wilk, ST John, Artem Artemev, James Hensman
We present a variational approximation for a wide range of GP models that does not require a matrix inverse to be performed at each optimisation step.
no code implementations • pproximateinference AABI Symposium 2022 • Mark van der Wilk, Artem Artemev, James Hensman
The need for matrix decompositions (inverses) is often named as a major impediment to scaling Gaussian process (GP) models, even in efficient approximations.
no code implementations • 15 Feb 2024 • Sebastian W. Ober, Artem Artemev, Marcel Wagenländer, Rudolfs Grobins, Mark van der Wilk
To address this, we make recommendations for comparing GP approximations based on a specification of what a user should expect from a method.