Global ETD Search

Search theses and dissertations gathered from participating repositories worldwide. Every result links back to the library that holds it. No account is needed.

Results

Showing 1 to 8 of 8 for “"input-dependent noise"”.

  1. Sequential learning, large-scale calibration, and uncertainty quantification

    … exploration to facilitate separating signal from input-dependent noise. Motivated by challenges in both large data size and model fidelity arising from ever larger modern computer experiments, highly accurate and computationally efficient divide-and-conquer calibration methods based on on-site …

    vt Repository record for Sequential learning, large-scale calibration, and uncertainty quantification (opens in a new tab)

  2. Consistency and Uniform Bounds for Heteroscedastic Simulation Metamodeling and Their Applications

    … provides an accurate approximation of the input-output relationship implied by a stochastic simulation experiment whose output is subject to input-dependent noise variance. Several challenges remain unsolved in this field. First, in-depth investigations into the consistency of …

    vt Repository record for Consistency and Uniform Bounds for Heteroscedastic Simulation Metamodeling and Their Applications (opens in a new tab)

  3. Heteroskedastic Gaussian processes for ecological forecasting

    … series share a common challenge: they exhibit input dependent noise. In such situations, accurate predictions into the future along with principled uncertainty are essential. Heteroskedastic Gaussian processes (hetGPs) are a flexible and nonparametric modeling technique wherein, two processes …

    vt Repository record for Heteroskedastic Gaussian processes for ecological forecasting (opens in a new tab)

  4. Kernel-Based Metamodeling for Heteroscedastic Simulation: Theory and Distribution-Free Uncertainty Quantification

    … characterizes near-optimal regimes, explains how noise-variance scaling governs worst-case performance, and proves high-probability pointwise bounds whose dimension dependence appears only through a log n factor. The third methodological piece is heteroscedastic weighted kernel ridge regression, …

    vt Repository record for Kernel-Based Metamodeling for Heteroscedastic Simulation: Theory and Distribution-Free Uncertainty Quantification (opens in a new tab)

  5. Data-Efficient Structure-Property Modeling with Heteroscedastic Gaussian Process Regression and Active Learning

    … materials data is often heteroscedastic, with input-dependent variability that complicates both surrogate prediction and data acquisition. This dissertation develops a Heteroscedastic Gaussian Process Regression (HGPR) framework integrated with active learning (AL) for data-efficient materials …

    jhu Repository record for Data-Efficient Structure-Property Modeling with Heteroscedastic Gaussian Process Regression and Active Learning (opens in a new tab)

  6. Computer Experimental Design for Gaussian Process Surrogates

    … and related fields is to efficiently explore the input space using a small number of samples, i.e., the experimental design problem. This dissertation focuses on the design problem under Gaussian process surrogates. The first work demonstrates empirically that space-filling designs disappoint when …

    vt Repository record for Computer Experimental Design for Gaussian Process Surrogates (opens in a new tab)

  7. Convex optimization for covariance steering and density control

    … for linear systems subjected to state and input-dependent noise processes. The terminal covariance assignment constraints are initially relaxed into semi-definite constraints. The solution to the relaxed problem is subsequently used to address the exact problem, demonstrating that this …

    texas Repository record for Convex optimization for covariance steering and density control (opens in a new tab)

  8. Gaussian Process Regression with Heteroscedastic Residuals and Fast MCMC Methods

    … for all observations. However, applications with input-dependent noise (heteroscedastic residuals) frequently arise in practice, as do applications in which the residuals do not have a Gaussian distribution. In this thesis, we propose a GP regression model with a latent variable that serves as an …

    toronto-retro Repository record for Gaussian Process Regression with Heteroscedastic Residuals and Fast MCMC Methods (opens in a new tab)