Hierarchical Gaussian Process Priors for Bayesian Neural Network Weights

Authors: Theofanis Karaletsos, Thang D. Bui

NeurIPS 2020 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental In this section, we evaluate the proposed priors and inference scheme on several regression and classification datasets, and study effects of per-datapoint priors as we proposed on extrapolation, interpolation, and out-of-distribution data.
Researcher Affiliation Collaboration Theofanis Karaletsos Facebook theokara@fb.com Thang D. Bui Uber AI and University of Sydney thang.bui@uber.com
Pseudocode No The paper does not contain any pseudocode or clearly labeled algorithm blocks.
Open Source Code No The paper states 'Additional updates and results will be available on https://arxiv.org/abs/2002.04033.' which is a link to the arXiv paper itself, not source code.
Open Datasets Yes We first illustrate the performance of the proposed model on a classification example. We generate a dataset of 100 data points and four classes
Dataset Splits No The paper mentions 'training points' and 'test sets' but does not specify explicit train/validation/test dataset splits with percentages, sample counts, or clear methodologies for reproduction.
Hardware Specification No The paper does not explicitly describe the hardware used for running its experiments, such as specific GPU/CPU models or memory details.
Software Dependencies No The paper does not provide specific software dependencies with version numbers.
Experiment Setup Yes We use M = 50 inducing weights for all experiments in this section. Details for the experimental settings and additional experiments are included in the appendices.