arXiv subjects

F. Bachoc

Publications and source records attributed to F. Bachoc.

4 recordsLinked to original sources

arXiv2024

Block-Additive Gaussian Processes under Monotonicity Constraints

We generalize the additive constrained Gaussian process framework to handle interactions between input variables while enforcing monotonicity constraints everywhere on the input space. The block-additive structure of the model is particularly suitable in the presence of interactions, while maintaining tractable computations. In addition, we develop a sequential algorithm, MaxMod, for model selection (i.e., the choice of the active input variables and of the blocks). We speed up our implementations through efficient matrix computations and thanks to explicit expressions of criteria involved in MaxMod. The performance and scalability of our methodology are showcased with several numerical examples in dimensions up to 120, as well as in a 5D real-world coastal flooding application, where interpretability is enhanced by the selection of the blocks.

stat.ME↗

arXiv2020

Bayesian Regression and Classification Using Gaussian Process Priors Indexed by Probability Density Functions

In this paper, we introduce the notion of Gaussian processes indexed by probability density functions for extending the Mat\'ern family of covariance functions. We use some tools from information geometry to improve the efficiency and the computational aspects of the Bayesian learning model. We particularly show how a Bayesian inference with a Gaussian process prior (covariance parameters estimation and prediction) can be put into action on the space of probability density functions. Our framework has the capacity of classifiying and infering on data observations that lie on nonlinear subspaces. Extensive experiments on multiple synthetic, semi-synthetic and real data demonstrate the effectiveness and the efficiency of the proposed methods in comparison with current state-of-the-art methods.

stat.ME↗

arXiv2020

Multioutput Gaussian Processes with Functional Data: A Study on Coastal Flood Hazard Assessment

Surrogate models are often used to replace costly-to-evaluate complex coastal codes to achieve substantial computational savings. In many of those models, the hydrometeorological forcing conditions (inputs) or flood events (outputs) are conveniently parameterized by scalar representations, neglecting that the inputs are actually time series and that floods propagate spatially inland. Both facts are crucial in flood prediction for complex coastal systems. Our aim is to establish a surrogate model that accounts for time-varying inputs and provides information on spatially varying inland flooding. We introduce a multioutput Gaussian process model based on a separable kernel that correlates both functional inputs and spatial locations. Efficient implementations consider tensor-structured computations or sparse-variational approximations. In several experiments, we demonstrate the versatility of the model for both learning maps and inferring unobserved maps, numerically showing the convergence of predictions as the number of learning maps increases. We assess our framework in a coastal flood prediction application. Predictions are obtained with small error values within computation time highly compatible with short-term forecast requirements (on the order of minutes compared to the days required by hydrodynamic simulators). We conclude that our framework is a promising approach for forecast and early-warning systems.

stat.ML↗

arXiv2015

Asymptotic properties of multivariate tapering for estimation and prediction

Parameter estimation for and prediction of spatially or spatio--temporally correlated random processes are used in many areas and often require the solution of a large linear system based on the covariance matrix of the observations. In recent years, the dataset sizes to which these methods are applied have steadily increased such that straightforward statistical tools are computationally too expensive to be used. In the univariate context, tapering, i.e., creating sparse approximate linear systems, has been shown to be an efficient tool in both the estimation and prediction settings. The asymptotic properties are derived under an infill asymptotic setting. In this paper we use a domain increasing framework for estimation and prediction using multivariate tapering. Under this asymptotic regime we prove that tapering (one-tapered form) preserves the consistency of the untapered maximum likelihood estimator and show that tapering has asymptotically the same mean squared prediction error as using the corresponding untapered predictor. The theoretical results are illustrated with simulations.

math.ST↗