Scalable and Robust Bayesian Inference via the Median Posterior

Authors: Stanislav Minsker, Sanvesh Srivastava, Lizhen Lin, David Dunson

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

Reproducibility Variable Result LLM Response
Research Type Experimental 4. Numerical Experiments, 4.1. Simulated data, 4.2. Real data: Pd G hormone levels vs day of ovulation
Researcher Affiliation Academia Departments of Mathematics1 and Statistical Science2, Duke University, Durham, NC 27708 Statistical and Applied Mathematical Sciences Institute3, 19 T.W. Alexander Dr, Research Triangle Park, NC 27709
Pseudocode Yes Algorithm 1 Evaluating the geometric median of probability distributions via Weiszfeld s algorithm, Algorithm 2 Approximating the M-posterior distribution
Open Source Code No The paper does not contain any explicit statements or links indicating that the source code for the methodology described is publicly available.
Open Datasets Yes North Carolina Early Pregnancy Study (NCEPS) measured urinary pregnanediol-3-glucuronide (Pd G) levels, a progesterone metabolite, in 166 women from the day of ovulation across 41 time points (Baird et al., 1999).
Dataset Splits Yes The data was divided into 10 subsets. On each stage, 9 of them were used to evaluate the M-posterior while the remaining was a test set; the process was repeated 10 times for different test subsets.
Hardware Specification No The paper does not provide any specific details about the hardware used for running the experiments.
Software Dependencies No The gausspr function in kernlab R package (Karatzoglou et al., 2004) is used for GP regression. (No version numbers provided for R or kernlab)
Experiment Setup Yes the noise variance (or nugget effect) is fixed at 0.01., we set m = 10 and generate 100 samples from every Π100,10( |Gj,i), j = 1, . . . , 10 to form the empirical measures Qj,i