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 20 of 42 for “"Bayesian uncertainty"”.
-
Bayesian Uncertainty Quantification for Large Scale Spatial Inverse Problems
We considered a Bayesian approach to nonlinear inverse problems in which the unknown quantity is a high dimension spatial field. The Bayesian approach contains a natural mechanism for regularization in the form of prior information, can incorporate information from heterogeneous sources and …
-
Bayesian Uncertainty Quantification while Leveraging Multiple Computer Model Runs
… not apply. To develop an approach to quantify uncertainty of computer model output in this setting, we leverage information from a collection of computer model runs (e.g., historical forecast and observation pairs for tropical cyclone precipitation totals) through a Bayesian hierarchical …
-
NEURAL NETWORK SURROGATES FOR AIR DEFENSE SIMULATION WITH BAYESIAN UNCERTAINTY QUANTIFICATION
This thesis developed and evaluated Bayesian convolutional neural network (CNN) models to act as surrogates for high-fidelity Air Defense Artillery (ADA) simulations. These neural network models are intended for use at tactical command posts to inform decision-making on the placement of ADA assets …
-
Bayesian uncertainty quantification of computer models with efficient calibration and computation
… The computational and statistical tools of Uncertainty Quantification (UQ) are instrumental in assessing how accurately a computer model describes a physical process. Bayesian framework for UQ has become the dominant approach, because it provides a principled way of quantifying uncertainty …
-
Bayesian uncertainty quantification and optimal experimental design in data driven simulations of engineering systems
The objective of this thesis is twofold. First, Bayesian uncertainty quantification, propagation, and model selection theories are outlined and applied to real-world problems, and second, Bayesian optimal experimental design theories are studied from a theoretical point of view and also applied to …
-
Bayesian uncertainty quantification of Langmuir probe and emission measurements applied to non-equilibrium argon and nitrogen plasmas
… electron density and mean energy. Rigorous uncertainty quantification is applied throughout, using new Bayesian inversion frameworks to account for data, experimental and modeling uncertainties. The ten Ar($4p$) state population densities ($2p_{1-10}$ in Paschen notation) are found to range …
-
Plant-Wide Process Modeling, Techno-Economic Optimization, and Bayesian Uncertainty Quantification for Manufacturing Value-Added Products from Lignocellulosic Biomass
… due to several interacting sources of uncertainty that can have strong impact on process performance, product quality, production rate, and economic viability. These uncertainties may arise from feedstock composition, limited experimental data, uncertain reaction pathways and kinetic …
-
Quantifying Structural Uncertainty in Hydrologic Models
… due to imperfect model structures (structural uncertainty), parameter estimation challenges (parameter uncertainty), and limitations in observational data and model forcings (input uncertainty). Bayesian inference has become a widely used framework for quantifying these uncertainties because it …
-
Interpretable machine learning and governing law discovery
… identify governing equations, quantify uncertainty, and make reliable predictions from sparse, noisy, and partially observed measurements. In this thesis, I present my work toward developing machine learning methods for interpretable physical law discovery in realistic data regimes. In …
-
Adding Model Uncertainty to Depth Prediction
… could be inaccurate. In this work, we add Bayesian uncertainty to pretrained convolutional neural networks. Testing the networks on a synthetic dataset shows that the uncertainty is able to give confidence levels that are linked with the accuracy of the model output. Additionally, masking …
-
Bayesian predictive posterior distributions
Bayesian uncertainty can be characterized in a number of ways, the usual one starting with a prior distribution which represents prior uncertainty as to the value of a parameter. This gets updated to the posterior quantification of uncertainty via the data evidence using the likelihood function. …
-
Learning from Disparate Data: Applications in Biometrics and Sustainability
… Learning using Gaussian Processes which employs Bayesian uncertainty to combine multiple views, and, ii). Dis- criminative Factorized Subspace Learning, which produces a low-dimensional factorized subspace, consisting of shared and per-view components. We have shown the efficacy of our algorithms …
-
Analysis of New Methods for Inference in Markov Decision Processes
… for computing statistics for value functions in Bayesian reinforcement learning. This involves investigating a recently developed mean field theory and its relationship with expectations of maxima of sets of random variables. The objective is to precisely calculate the mean of posterior optimal …
-
Essays on reputation and information
… features a repeated auction that also involves Bayesian uncertainty, about the `type’ of the seller. In addition, we present the predictions of a theoretical model that extends the existing sequential equilibrium literature into multi-player, market-based games. We find that reputational …
-
Assessing the dynamic status of fish resources in space and time.
… season. Two models are presented based on Bayesian uncertainty statistical theory. A working computer decision model is developed to estimate the weekly stock abundance by area. A simulation model is developed to examine the strategic plan of the fishery. A visualized computer application …
-
Generative Models for Uncertainty of Medical and Seismic Imaging
… by non-unique solutions, we adopt a Bayesian framework. This framework integrates prior knowledge and observational data, resulting in a posterior distribution that characterizes the range of plausible solutions. Such uncertainty quantification is critical in applications where …
-
Forecasting lightpath quality of transmission and implementing uncertainty in the forecast models
… only predict the target but also to measure the uncertainty of the prediction accurately. Working with time series requires reliable forecasters. An intuitive N-Beats architecture was implemented in this thesis with the desirable properties of being fast to train and adaptable, without requiring …
Page 1 of 3