26,135 Hits in 5.2 sec

Multi-fidelity data fusion through parameter space reduction with applications to automotive engineering [article]

Francesco Romor, Marco Tezzele, Markus Mrosek, Carsten Othmer, Gianluigi Rozza
2022 arXiv   pre-print
Then we build a low-fidelity response surface based on such reduction, thus enabling multi-fidelity Gaussian process regression without the need of running new simulations with simplified physical models  ...  In the context of Gaussian process regression we can exploit low-fidelity models to better capture the latent manifold thus improving the accuracy of the model.  ...  multi-fidelity Gaussian process regression In this section we briefly present the nonlinear autoregressive multi-fidelity Gaussian process regression (NARGP) scheme proposed in [37] .  ... 
arXiv:2110.14396v2 fatcat:q7pioug5tvgezpw3ghwr6lf4pe

Multi-fidelity data fusion for the approximation of scalar functions with low intrinsic dimensionality through active subspaces [article]

Francesco Romor, Marco Tezzele, Gianluigi Rozza
2020 arXiv   pre-print
Gaussian processes are employed for non-parametric regression in a Bayesian setting.  ...  When the model's gradient information is provided, the presence of an active subspace can be exploited to design low-fidelity response surfaces and thus enable Gaussian process multi-fidelity regression  ...  multi-fidelity Gaussian process regression We adopt the nonlinear autoregressive multi-fidelity Gaussian process regression (NARGP) scheme proposed in [26] .  ... 
arXiv:2010.08349v1 fatcat:gsez6wmkkrg6xgyxs42mthehdy

Gaussian process regression with multiple response variables

Bo Wang, Tao Chen
2015 Chemometrics and Intelligent Laboratory Systems  
Gaussian process regression (GPR) is a Bayesian non-parametric technology that has gained extensive application in data-based modelling of various systems, including those of interest to chemometrics.  ...  In the paper we propose a direct formulation of the covariance function for multi-response GPR, based on the idea that its covariance function is assumed to be the "nominal" uni-output covariance multiplied  ...  Gaussian process regressions and the widely used partial least squares regression for multi-inputs and multi-outputs (PLS) are also performed, where the former is conducted for the two outputs independently  ... 
doi:10.1016/j.chemolab.2015.01.016 fatcat:dl6qwy4bzrelxfdfbfl5x5rvfa

Multivariate Gaussian and Student-t Process Regression for Multi-output Prediction [article]

Zexun Chen, Bo Wang, Alexander N. Gorban
2019 arXiv   pre-print
Gaussian process model for vector-valued function has been shown to be useful for multi-output prediction.  ...  In this paper, we propose a unified framework which is used not only to introduce a novel multivariate Student-t process regression model (MV-TPR) for multi-output prediction, but also to reformulate the  ...  Therefore, the existing methods for multi-output Gaussian process regression cannot be applied to Student−t process regression.  ... 
arXiv:1703.04455v6 fatcat:b5dgnjkb2fdsjdj27cpwl4r5bm

Information Agents for Pervasive Sensor Networks

Alex Rogers, Mike Osborne, Sarvapali D. Ramchurn, Stephen Roberts, Nicholas R Jennings
2008 2008 Sixth Annual IEEE International Conference on Pervasive Computing and Communications (PerCom)  
Gaussian process to build a probabilistic model of the environmental parameters being measured by local sensors, and the correlations and delays that exist between them.  ...  Our motivating scenario is the need to provide situational awareness support to first responders at the scene of a large scale incident, and we describe how we use an iterative formulation of a multi-output  ...  Figure 3 : 3 Prediction and regression of tide height data for (a) independent and (b) multi-output Gaussian processes.  ... 
doi:10.1109/percom.2008.22 dblp:conf/percom/RogersORRJ08 fatcat:zc7oqkiz7fefnfgeljfll2muwu

Remarks on multivariate Gaussian Process [article]

Zexun Chen, Jun Fan, Kuo Wang
2021 arXiv   pre-print
a useful statistical learning method for multi-output prediction problems.  ...  In this paper, we propose a precise definition of multivariate Gaussian processes based on Gaussian measures on vector-valued function spaces, and provide an existence proof.  ...  Multivariate Gaussian process regression As a useful application, multi-output prediction using multivariate Gaussian process is a good example.  ... 
arXiv:2010.09830v3 fatcat:rremwhn4ibalnpq5dcgtsgoud4

Scalable Inference for Gaussian Process Models with Black-Box Likelihoods

Amir Dezfouli, Edwin V. Bonilla
2015 Neural Information Processing Systems  
Experiments on small datasets for various problems including regression, classification, Log Gaussian Cox processes, and warped GPs show that our method can perform as well as the full method under high  ...  We propose a sparse method for scalable automated variational inference (AVI) in a large class of models with Gaussian process (GP) priors, multiple latent functions, multiple outputs and non-linear likelihoods  ...  Gaussian process regression networks on the SARCOS dataset.  ... 
dblp:conf/nips/DezfouliB15 fatcat:fhx5627irzdm7imeppau36lcry

A probabilistic data-driven model for planar pushing [article]

Maria Bauza, Alberto Rodriguez
2017 arXiv   pre-print
The learned models rely on a variation of Gaussian processes with input-dependent noise called Variational Heteroscedastic Gaussian processes (VHGP) that capture the mean and variance of a stochastic function  ...  These algorithms are referred as heteroscedastic Gaussian processes (HGPs) and can regress both the mean of the process and its variance over the input space.  ...  . · Model: The learned model is based on a family of Gaussian processes called Heteroscedastic Gaussian processes (HGPs), along with their state-of-the-art variational implementation [1] .  ... 
arXiv:1704.03033v2 fatcat:du3v2t5kzrajjj5wlooirpixmm

Benchmarking Regression Methods: A comparison with CGAN [article]

Karan Aggarwal, Matthieu Kirchmeyer, Pranjul Yadav, S. Sathiya Keerthi, Patrick Gallinari
2020 arXiv   pre-print
The standard approach to solve regression problems is to probabilistically model the output y as the sum of a mean function m(x) and a noise term z; it is also usual to take the noise to be a Gaussian.  ...  Excellent solutions have been demonstrated mostly in image processing applications which involve large, continuous output spaces.  ...  Design of scalable CGAN implementations suited for regression is another key direction. Designing a Bayesian version of CGAN [19] would give great gains on small training datasets.  ... 
arXiv:1905.12868v5 fatcat:l4jet3kihne3vetwqpzdvcqw7a

Multi-fidelity surrogate modeling for time-series outputs [article]

Baptiste Kerleguer
2022 arXiv   pre-print
Using an experimental design of the low-and high-fidelity code levels, an original Gaussian process regression method is proposed.  ...  The code output is expanded on a basis built from the experimental design. The first coefficients of the expansion of the code output are processed by a co-kriging approach.  ...  Gaussian process regression for functional outputs. In this subsection, we address Gaussian process regression for a simple-fidelity code with time-series output.  ... 
arXiv:2109.11374v2 fatcat:jectsrsqcja43clq3l5q7ytiqi

Multivariate Gaussian and Student-t process regression for multi-output prediction

Zexun Chen, Bo Wang, Alexander N. Gorban
2019 Neural computing & applications (Print)  
Gaussian process model for vector-valued function has been shown to be useful for multi-output prediction.  ...  In this paper, we propose a unified framework which is used not only to introduce a novel multivariate Student-t process regression model (MV-TPR) for multi-output prediction, but also to reformulate the  ...  Therefore, the existing methods for multi-output Gaussian process regression cannot be applied to Student-t process regression.  ... 
doi:10.1007/s00521-019-04687-8 fatcat:qg55jv7q7jf7vdmyromcf7tix4

Coregionalised Locomotion Envelopes - A Qualitative Approach [article]

Neil Dhir, Houman Dallali, Mo Rastgaar
2018 arXiv   pre-print
In this short paper we introduce coregionalised locomotion envelopes - a method for multi-dimensional manifold regression, on human locomotion variates.  ...  'Sharing of statistical strength' is a phrase often employed in machine learning and signal processing.  ...  Coregionalised Gaussian processes The coregionalised regression model relies upon the use of vector-valued kernels, one of the most common approaches for this regression is the linear model of coregionalisation  ... 
arXiv:1803.04965v1 fatcat:ze5lxdyecbhcdndza7khy674t4

Dirichlet-based Gaussian Processes for Large-scale Calibrated Classification [article]

Dimitrios Milios, Raffaello Camoriano, Pietro Michiardi, Lorenzo Rosasco, Maurizio Filippone
2018 arXiv   pre-print
In this work, we investigate if and how Gaussian process regression directly applied to the classification labels can be used to tackle this question.  ...  To this aim, we propose a novel approach based on interpreting the labels as the output of a Dirichlet distribution.  ...  Remark: In the binary classification case, we still have to perform regression on two latent processes.  ... 
arXiv:1805.10915v1 fatcat:rl3mr53fp5cqppczvg4yuacnvu

A Multiway Gaussian Mixture Model based Adaptive Kernel Partial Least Squares Regression Approach for Inferential Quality Predictions of Batch Processes

Jie Yu
2012 IFAC Proceedings Volumes  
Soft sensor technique has become increasingly important to provide reliable on-line measurements, facilitate advanced process control and improve product quality in process industries.  ...  Then the multiway Gaussian mixture model is estimated with multiple Gaussian clusters in the kernel space.  ...  Finally, the concluding remarks are drawn in Section V.  ... 
doi:10.3182/20120710-4-sg-2026.00086 fatcat:s45fsgbthjborojobo5ebc7fhm

Aerodynamic probe calibration using Gaussian process regression

Florian Marco Heckmeier, Christian Breitsamter
2020 Measurement science and technology  
With the help of statistical methods, more precisely Gaussian process regressions, this similarity is exploited in order to use existing calibration data of different probes reducing the calibration time  ...  The number of calibration points in the five-hole probe case is reduced by at least one order of magnitude with comparable reconstruction accuracy.  ...  Acknowledgments The authors would like to thank the company Vectoflow GmbH for providing the multi-hole pressure probe calibration data of numerous different probes.  ... 
doi:10.1088/1361-6501/aba37d fatcat:33wumttfovhtbef5t3kdoibxoa
« Previous Showing results 1 — 15 out of 26,135 results