# Sequential construction and dimension reduction of Gaussian processes under inequality constraints

@article{Bachoc2020SequentialCA, title={Sequential construction and dimension reduction of Gaussian processes under inequality constraints}, author={F. Bachoc and A. F. Lopera and O. Roustant}, journal={arXiv: Statistics Theory}, year={2020} }

Accounting for inequality constraints, such as boundedness, monotonicity or convexity, is challenging when modeling costly-to-evaluate black box functions. In this regard, finite-dimensional Gaussian process (GP) models bring a valuable solution, as they guarantee that the inequality constraints are satisfied everywhere. Nevertheless, these models are currently restricted to small dimensional situations (up to dimension 5). Addressing this issue, we introduce the MaxMod algorithm that… Expand

#### References

SHOWING 1-10 OF 46 REFERENCES

Finite-dimensional Gaussian approximation with linear inequality constraints

- Mathematics, Computer Science
- SIAM/ASA J. Uncertain. Quantification
- 2018

The finite-dimensional Gaussian approach from Maatouk and Bay (2017) is considered which can satisfy inequality conditions everywhere and its full framework together with a Hamiltonian Monte Carlo-based sampler provides efficient results on both data fitting and uncertainty quantification. Expand

Gaussian process regression with linear inequality constraints

- Computer Science, Mathematics
- Reliab. Eng. Syst. Saf.
- 2020

A framework for incorporating any type of linear constraints in Gaussian process modeling is introduced, this including common bound and monotonicity constraints, and a sequential sampling strategy is developed where the input space is explored via a criterion which maximizes the probability of respecting the given constraints. Expand

Bounded Regression with Gaussian Process Projection

- Mathematics
- 2018

Examples with bound information on the regression function and density abound in many real applications. We propose a novel approach for estimating such functions by incorporating the prior knowledge… Expand

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

- Computer Science, Mathematics
- ArXiv
- 2019

A noise term for the relaxation of the interpolation conditions is introduced, and the corresponding approximation of GP emulators under linear inequality constraints is developed, showing that the performance of MC and MCMC samplers improves when considering noisy observations. Expand

Finite-dimensional approximation of Gaussian processes with inequality constraints

- Mathematics
- 2017

Due to their flexibility, Gaussian processes (GPs) have been widely used in nonparametric function estimation. A prior information about the underlying function is often available. For instance, the… Expand

Gaussian process modeling with inequality constraints

- Computer Science
- 2012

This paper introduces a new framework for incorporating constraints in Gaussian process modeling, including bound, monotonicity and convexity constraints, and extends this framework to any type of linear constraint. Expand

Gaussian Process Modulated Cox Processes under Linear Inequality Constraints

- Mathematics, Computer Science
- AISTATS
- 2019

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 is introduced. Expand

Kriging of financial term-structures

- Mathematics, Computer Science
- Eur. J. Oper. Res.
- 2016

A new term-structure interpolation method that extends classical spline techniques by additionally allowing for quantification of uncertainty by based on a generalization of kriging models with linear equality constraints (market-fit conditions) and shape-preserving conditions such as monotonicity or positivity (no-arbitrage conditions). Expand

Maximum likelihood estimation for Gaussian processes under inequality constraints

- Mathematics
- Electronic Journal of Statistics
- 2019

We consider covariance parameter estimation for a Gaussian process under inequality constraints (boundedness, monotonicity or convexity) in fixed-domain asymptotics. We address the estimation of the… Expand

A supermartingale approach to Gaussian process based sequential design of experiments

- Mathematics
- Bernoulli
- 2019

Gaussian process (GP) models have become a well-established frameworkfor the adaptive design of costly experiments, and notably of computerexperiments. GP-based sequential designs have been found… Expand