Stein Point Markov Chain Monte Carlo

Authors: Wilson Ye Chen, Alessandro Barp, Francois-Xavier Briol, Jackson Gorham, Mark Girolami, Lester Mackey, Chris Oates

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

Reproducibility Variable Result LLM Response
Research Type Experimental In this section our attention turns to the empirical performance of SP-MCMC. The experimental protocol is explained in Section 4.1 and specific experiments are described in Sections 4.2, 4.3 and 4.4.
Researcher Affiliation Collaboration 1Institute of Statistical Mathematics 2Imperial College London 3Alan Turing Institute 4University of Cambridge 5Open Door 6Microsoft Research 7Newcastle University.
Pseudocode No The paper describes the method in paragraph form but does not provide structured pseudocode or an algorithm block.
Open Source Code Yes Code to reproduce all experiments can be downloaded at https://github.com/wilson-ye-chen/sp-mcmc.
Open Datasets No The paper describes using 2,000 daily percentage returns of the S&P 500 stock index for the IGARCH model and data for an ODE model, but does not provide explicit links, DOIs, repositories, or formal citations for public access to these datasets.
Dataset Splits No The paper does not provide specific dataset split information (percentages, sample counts, or methodology) for training, validation, or testing.
Hardware Specification No The paper does not provide specific hardware details such as GPU or CPU models, processor types, or memory used for running the experiments.
Software Dependencies No The paper mentions algorithms and kernels used (e.g., IMQ kernel, RWM, MALA, MED, SVGD) but does not specify any software dependencies with version numbers (e.g., Python version, specific library versions like PyTorch, TensorFlow, or NumPy).
Experiment Setup Yes For SP-MCMC the sequence (mj)j N was set as mj = 5. Results are presented in Fig. 2 with n = 1000... SP-MCMC was implemented with mj = 5j... All methods produced a point set of size n = 1000... SP-MCMC was implemented with the INFL criterion and mj = 10 (d = 4), mj = 20 (d = 10).