A Generalized Gaussian Process Model for Computer Experiments With Binary Time Series

@article{Sung2019AGG,
  title={A Generalized Gaussian Process Model for Computer Experiments With Binary Time Series},
  author={Chih-Li Sung and Ying Hung and William Rittase and Cheng Zhu and C. F. Jeff Wu},
  journal={Journal of the American Statistical Association},
  year={2019},
  volume={115},
  pages={945 - 956}
}
Abstract Non-Gaussian observations such as binary responses are common in some computer experiments. Motivated by the analysis of a class of cell adhesion experiments, we introduce a generalized Gaussian process model for binary responses, which shares some common features with standard GP models. In addition, the proposed model incorporates a flexible mean function that can capture different types of time series structures. Asymptotic properties of the estimators are derived, and an optimal… Expand
Latent function-on-scalar regression models for observed sequences of binary data: a restricted likelihood approach
TLDR
This paper proposes a practical computational framework for maximum likelihood analysis via the parameter expansion technique, and relies on the use of a complete data likelihood, with the advantage of being able to handle non-equally spaced and missing observations effectively. Expand
Renewal model for dependent binary sequences
We suggest to construct infinite stochastic binary sequences by associating one of the two symbols of the sequence with the renewal times of an underlying renewal process. Focusing on stationaryExpand
Modelling Function-Valued Processes with Nonseparable Covariance Structure
We discuss a general Bayesian framework on modelling multidimensional function-valued processes by using a Gaussian process or a heavy-tailed process as a prior, enabling us to handle nonseparableExpand
Calibration for Computer Experiments With Binary Responses and Application to Cell Adhesion Study
TLDR
A new calibration framework for binary responses is proposed for T cell adhesion experiments, which provides insight into the unknown values of the kinetic parameters which are difficult to determine by physical experiments due to the limitation of the existing experimental techniques. Expand
On I-Optimal Designs for Generalized Linear Models: An Efficient Algorithm via General Equivalence Theory
The generalized linear model plays an important role in statistical analysis and the related design issues are undoubtedly challenging. The state-of-the-art works mostly apply to design criteria onExpand
An efficient algorithm for Elastic I‐optimal design of generalized linear models
TLDR
This work considers the Elastic I-optimality as a prediction-oriented design criterion for generalized linear models, and develops efficient algorithms for such $\text{EI}$-optimal designs, and proposes an efficient algorithm that adequately combines the Fedorov-Wynn algorithm and multiplicative algorithm. Expand
Sequential Design of Computer Experiments with Quantitative and Qualitative Factors in Applications to HPC Performance Optimization
TLDR
The proposed adaptive CEE method combines the predictive mean and standard deviation based on the additive Gaussian process to achieve a meaningful balance between exploitation and exploration for optimization of computer experiments with qualitative and quantitative factors. Expand
Ice Model Calibration Using Semi-continuous Spatial Data
TLDR
A hierarchical latent variable model is introduced that handles binary spatial patterns and positive continuous spatial patterns as separate components and is applied to calibrate a physical model for the Antarctic ice sheet and obtains improved future ice-volume change projections. Expand
Computer Model Calibration with Time Series Data using Deep Learning and Quantile Regression
TLDR
This work proposes a new calibration framework based on a deep neural network (DNN) with long-short term memory layers that directly emulates the inverse relationship between the model output and input parameters that can yield accurate point estimates and well calibrated interval estimates for input parameters. Expand
Analysis of multivariate non-gaussian functional data: A semiparametric latent process approach
  • Jiakun Jiang, Huazhen Lin, Qingzhi Zhong, Yi Li
  • Journal of Multivariate Analysis
  • 2021
...
1
2
...

References

SHOWING 1-10 OF 86 REFERENCES
Generalized Gaussian Process Regression Model for Non-Gaussian Functional Data
  • Bo Wang, J. Shi
  • Mathematics
  • 2014
In this article, we propose a generalized Gaussian process concurrent regression model for functional data, where the functional response variable has a binomial, Poisson, or other non-GaussianExpand
Local Gaussian Process Approximation for Large Computer Experiments
TLDR
A family of local sequential design schemes that dynamically define the support of a Gaussian process predictor based on a local subset of the data are derived, enabling a global predictor able to take advantage of modern multicore architectures. Expand
Markov regression models for time series: a quasi-likelihood approach.
TLDR
A quasi-likelihood (QL) approach to regression analysis with time series data is discussed, analogous to QL for independent observations, large-sample properties of the regression coefficients depend only on correct specification of the first conditional moment. Expand
Approximations for Binary Gaussian Process Classification
We provide a comprehensive overview of many recent algorithms for approximate inference in Gaussian process models for probabilistic binary classification. The relationships between severalExpand
Generalized Autoregressive Moving Average Models
A class of generalized autoregressive moving average (GARMA) models is developed that extends the univariate Gaussian ARMA time series model to a flexible observation-driven model for non-GaussianExpand
Binary Time Series Modeling With Application to Adhesion Frequency Experiments
TLDR
A binary time series model incorporating random effects is developed and a goodness-of-fit statistic is introduced to assess the adequacy of distribution assumptions on the dependent binary data with random effects. Expand
Orthogonal Gaussian process models
TLDR
A new Gaussian process model whose stochastic part is orthogonal to the mean part to address Identifiability issues and applications to multi-fidelity simulations using data examples are discussed. Expand
Particle Learning of Gaussian Process Models for Sequential Design and Optimization
We develop a simulation-based method for the online updating of Gaussian process regression and classification models. Our method exploits sequential Monte Carlo to produce a fast sequential designExpand
A dynamic modelling strategy for Bayesian computer model emulation
TLDR
Bayesian multivariate dynamic linear models with Gaussian process modelling are combined in an effective manner, and the general strategy will be useful for other computer model evaluation studies with time series or functional outputs. Expand
Multivariate Gaussian Process Emulators With Nonseparable Covariance Structures
TLDR
Nonseparable covariance structures for Gaussian process emulators are developed, based on the linear model of coregionalization and convolution methods, finding that only emulators with nonseparable covariances structures have sufficient flexibility both to give good predictions and to represent joint uncertainty about the simulator outputs appropriately. Expand
...
1
2
3
4
5
...