Practical and Rigorous Uncertainty Bounds for Gaussian Process Regression

Authors: Christian Fiedler, Carsten W. Scherer, Sebastian Trimpe7439-7447

AAAI 2021 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We now test the theoretical results in numerical experiments using synthetic data, where the ground truth is known. First, we investigate the frequentist behaviour of the uncertainty bounds.
Researcher Affiliation Academia Christian Fiedler,1, 2, 3 Carsten W. Scherer,2 Sebastian Trimpe 1, 3 1Intelligent Control Systems Group, Max Planck Institute for Intelligent Systems 2Mathematical Systems Theory, University of Stuttgart 3Institute for Data Science in Mechanical Engineering, RWTH Aachen University
Pseudocode No The paper does not contain any structured pseudocode or algorithm blocks.
Open Source Code No The paper does not explicitly state that the code for the methodology described in this paper is open-source or publicly available.
Open Datasets No The paper uses synthetic data generated according to described processes (e.g., 'generate randomly 50 functions (the ground truths) from an RKHS', 'randomly generate the function r...from the RKHS'), but does not provide concrete access information (link, DOI, repository) for a publicly available or open dataset.
Dataset Splits No The paper describes generating training data by sampling inputs and evaluating results on an 'equidistant grid', but does not specify explicit training/validation/test dataset splits with percentages or sample counts.
Hardware Specification No The paper does not provide specific hardware details such as exact GPU/CPU models, processor types, or memory amounts used for running its experiments.
Software Dependencies No The paper does not provide specific ancillary software details with version numbers (e.g., library or solver names with version numbers) needed to replicate the experiment.
Experiment Setup Yes Unless noted otherwise, for each of the following experiments we use D = [ 1, 1] as input space and generate randomly 50 functions (the ground truths) from an RKHS and evaluate these on a grid 1000 equidistant points from D. ... We sample uniformly 50 inputs, evaluate the ground truth on these inputs and adding normal zero-mean i.i.d. noise with standard deviation (SD) 0.5.