Semi-Separable Hamiltonian Monte Carlo for Inference in Bayesian Hierarchical Models

Authors: Yichuan Zhang, Charles Sutton

NeurIPS 2014 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental In this section, we compare the performance of SSHMC with the standard HMC and RMHMC within Gibbs [7] on four benchmark models. The performance is evaluated by the minimum Effective Sample Size (ESS) over all dimensions (see [6]).
Researcher Affiliation Academia Yichuan Zhang School of Informatics University of Edinburgh Y.Zhang-60@sms.ed.ac.uk Charles Sutton School of Informatics University of Edinburgh c.sutton@inf.ed.ac.uk
Pseudocode Yes Algorithm 1 SSHMC by ABLA
Open Source Code No The paper does not provide an explicit statement about the release of source code for the described methodology, nor does it include a link to a code repository.
Open Datasets Yes We use the Statlog (German credit) dataset from [1]. [1] K. Bache and M. Lichman. UCI machine learning repository, 2013. URL http://archive.ics. uci.edu/ml.
Dataset Splits No The paper discusses tuning parameters like step size and number of leapfrog steps, but it does not specify explicit training, validation, or test dataset splits in terms of percentages or counts for reproducibility.
Hardware Specification No The paper does not provide specific hardware details such as GPU/CPU models, processor types, or memory used for running the experiments. It does not mention any specific computing environments beyond general descriptions.
Software Dependencies No The paper does not provide specific software dependency details, such as library names with version numbers (e.g., Python 3.x, PyTorch 1.x).
Experiment Setup Yes The step size of all methods are manually tuned so that the acceptance rate is around 70-85%. The number of leapfrog steps are tuned for each method using preliminary runs. We use 2 leapfrog steps for low-level parameters and 1 leapfrog step for the hyperparameter in ABLA and the same leapfrog step size for the two separable Hamiltonians.