Minimum Stein Discrepancy Estimators

Authors: Alessandro Barp, Francois-Xavier Briol, Andrew Duncan, Mark Girolami, Lester Mackey

NeurIPS 2019 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We illustrate this advantage for several challenging problems for score matching, such as non-smooth, heavy-tailed or light-tailed densities.
Researcher Affiliation Collaboration Alessandro Barp Department of Mathematics Imperial College London a.barp16@imperial.ac.uk; François-Xavier Briol Department of Statistical Science University College London f.briol@ucl.ac.uk; Andrew B. Duncan Department of Mathematics Imperial College London a.duncan@imperial.ac.uk; Mark Girolami Department of Engineering University of Cambridge mag92@eng.cam.ac.uk; Lester Mackey Microsoft Research Cambridge, MA, USA lmackey@microsoft.com
Pseudocode No The paper describes algorithms and derivations in prose and mathematical notation but does not include a clearly labeled pseudocode block or algorithm.
Open Source Code No The paper does not contain any explicit statement about releasing source code or provide a link to a code repository.
Open Datasets No The numerical experiments are conducted on data generated from specified distributions (e.g., symmetric Bessel distributions, non-standardised student-t distributions, generalised Gamma distributions). The paper does not provide access information (link, citation with author/year, or mention of standard public datasets) for any pre-existing, publicly available datasets.
Dataset Splits No The paper does not specify any train/validation/test dataset splits, sample counts for splits, or reference predefined splits for reproducibility.
Hardware Specification No The paper does not specify any hardware details (e.g., GPU models, CPU types, memory) used for running the experiments.
Software Dependencies No The paper does not list any specific software dependencies or their version numbers (e.g., 'Python 3.8', 'PyTorch 1.9') required to reproduce the experiments.
Experiment Setup Yes In the numerical experiments, specific details are provided such as 'Both algorithms use constant stepsizes and minibatches of size 50.' (Section 4.2) and 'We set n = 300 and corrupt 80 points by setting them to the value x = 8.' (Section 4.3).