Search Results for author: Nicolas Durrande

Found 20 papers, 5 papers with code

Kernel Identification Through Transformers

1 code implementation NeurIPS 2021 Fergus Simpson, Ian Davies, Vidhi Lalchand, Alessandro Vullo, Nicolas Durrande, Carl Rasmussen

Kernel selection plays a central role in determining the performance of Gaussian Process (GP) models, as the chosen kernel determines both the inductive biases and prior support of functions under the GP prior.

Deep Neural Networks as Point Estimates for Deep Gaussian Processes

no code implementations NeurIPS 2021 Vincent Dutordoir, James Hensman, Mark van der Wilk, Carl Henrik Ek, Zoubin Ghahramani, Nicolas Durrande

This results in models that can either be seen as neural networks with improved uncertainty prediction or deep Gaussian processes with increased prediction accuracy.

Bayesian Inference Gaussian Processes

The Minecraft Kernel: Modelling correlated Gaussian Processes in the Fourier domain

no code implementations11 Mar 2021 Fergus Simpson, Alexis Boukouvalas, Vaclav Cadek, Elvijs Sarkans, Nicolas Durrande

In the univariate setting, using the kernel spectral representation is an appealing approach for generating stationary covariance functions.

Gaussian Processes

A Tutorial on Sparse Gaussian Processes and Variational Inference

no code implementations27 Dec 2020 Felix Leibfried, Vincent Dutordoir, ST John, Nicolas Durrande

In this context, a convenient choice for approximate inference is variational inference (VI), where the problem of Bayesian inference is cast as an optimization problem -- namely, to maximize a lower bound of the log marginal likelihood.

Bayesian Inference Gaussian Processes +1

Matérn Gaussian Processes on Graphs

no code implementations29 Oct 2020 Viacheslav Borovitskiy, Iskander Azangulov, Alexander Terenin, Peter Mostowsky, Marc Peter Deisenroth, Nicolas Durrande

Gaussian processes are a versatile framework for learning unknown functions in a manner that permits one to utilize prior information about their properties.

Gaussian Processes

Sparse Gaussian Processes with Spherical Harmonic Features

no code implementations ICML 2020 Vincent Dutordoir, Nicolas Durrande, James Hensman

We introduce a new class of inter-domain variational Gaussian processes (GP) where data is mapped onto the unit hypersphere in order to use spherical harmonic representations.

Gaussian Processes

Regret Bounds for Noise-Free Bayesian Optimization

no code implementations12 Feb 2020 Sattar Vakili, Victor Picheny, Nicolas Durrande

Bayesian optimisation is a powerful method for non-convex black-box optimization in low data regimes.

Bayesian Optimisation

Doubly Sparse Variational Gaussian Processes

no code implementations15 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.

Gaussian Processes

Bayesian Quantile and Expectile Optimisation

no code implementations12 Jan 2020 Léonard Torossian, Victor Picheny, Nicolas Durrande

In this paper, we propose new variational models for Bayesian quantile and expectile regression that are well-suited for heteroscedastic settings.

Bayesian Optimisation Gaussian Processes

Gaussian Process Modulated Cox Processes under Linear Inequality Constraints

no code implementations28 Feb 2019 Andrés F. López-Lopera, ST John, Nicolas Durrande

We introduce a novel finite approximation of GP-modulated Cox processes where positiveness conditions can be imposed directly on the GP, with no restrictions on the covariance function.

Point Processes

Banded Matrix Operators for Gaussian Markov Models in the Automatic Differentiation Era

no code implementations26 Feb 2019 Nicolas Durrande, Vincent Adam, Lucas Bordeaux, Stefanos Eleftheriadis, James Hensman

Banded matrices can be used as precision matrices in several models including linear state-space models, some Gaussian processes, and Gaussian Markov random fields.

Gaussian Processes Variational Inference

Approximating Gaussian Process Emulators with Linear Inequality Constraints and Noisy Observations via MC and MCMC

no code implementations15 Jan 2019 Andrés F. López-Lopera, François Bachoc, Nicolas Durrande, Jérémy Rohmer, Déborah Idier, Olivier Roustant

Finally, on 2D and 5D coastal flooding applications, we show that more flexible and realistic GP implementations can be obtained by considering noise effects and by enforcing the (linear) inequality constraints.

Gaussian Processes

Scalable GAM using sparse variational Gaussian processes

no code implementations28 Dec 2018 Vincent Adam, Nicolas Durrande, ST John

Generalized additive models (GAMs) are a widely used class of models of interest to statisticians as they provide a flexible way to design interpretable models of data beyond linear models.

Additive models Gaussian Processes +1

Physically-Inspired Gaussian Process Models for Post-Transcriptional Regulation in Drosophila

1 code implementation29 Aug 2018 Andrés F. López-Lopera, Nicolas Durrande, Mauricio A. Alvarez

Since the post-transcriptional regulation of Drosophila depends on spatiotemporal interactions between mRNAs and gap proteins, proper physically-inspired stochastic models are required to study the link between both quantities.

Gaussian Processes

Finite-dimensional Gaussian approximation with linear inequality constraints

1 code implementation20 Oct 2017 Andrés F. López-Lopera, François Bachoc, Nicolas Durrande, Olivier Roustant

Introducing inequality constraints in Gaussian process (GP) models can lead to more realistic uncertainties in learning a great variety of real-world problems.

Variational Fourier features for Gaussian processes

1 code implementation21 Nov 2016 James Hensman, Nicolas Durrande, Arno Solin

This work brings together two powerful concepts in Gaussian processes: the variational approach to sparse approximation and the spectral representation of Gaussian processes.

Gaussian Processes

Nested Kriging predictions for datasets with large number of observations

1 code implementation19 Jul 2016 Didier Rullière, Nicolas Durrande, François Bachoc, Clément Chevalier

This work falls within the context of predicting the value of a real function at some input locations given a limited number of observations of this function.

An analytic comparison of regularization methods for Gaussian Processes

no code implementations2 Feb 2016 Hossein Mohammadi, Rodolphe Le Riche, Nicolas Durrande, Eric Touboul, Xavier Bay

A measure for data-model discrepancy is proposed which serves for choosing a regularization technique. In the second part of the paper, a distribution-wise GP is introduced that interpolates Gaussian distributions instead of data points.

Gaussian Processes

Invariances of random fields paths, with applications in Gaussian Process Regression

no code implementations6 Aug 2013 David Ginsbourger, Olivier Roustant, Nicolas Durrande

We study pathwise invariances of centred random fields that can be controlled through the covariance.

Cannot find the paper you are looking for? You can Submit a new open access paper.