Stein Points
Authors: Wilson Ye Chen, Lester Mackey, Jackson Gorham, Francois-Xavier Briol, Chris Oates
ICML 2018 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Our empirical results demonstrate that Stein Points enable accurate approximation of the posterior at modest computational cost. In addition, theoretical results are provided to establish convergence of the method. |
| Researcher Affiliation | Collaboration | 1School of Mathematical and Physical Sciences, University of Technology Sydney, Australia 2Microsoft Research New England, USA 3Opendoor Labs, Inc., USA 4Department of Statistics, University of Warwick, UK 5Department of Mathematics, Imperial College London, UK 6Alan Turing Institute, UK 7School of Mathematics, Statistics and Physics, Newcastle University, UK. |
| Pseudocode | No | The paper describes algorithms in prose and mathematical equations but does not include a formally labeled pseudocode or algorithm block. |
| Open Source Code | No | The paper does not include an unambiguous statement about releasing source code for the described methodology or provide a direct link to a code repository. |
| Open Datasets | No | The paper references data from a 'LIDAR experiment (Ruppert et al., 2003)' and 'S&P 500 stock index' but does not provide specific access information (links, repositories, or formal citations for public availability) for these datasets. |
| Dataset Splits | No | The paper does not provide specific details on dataset splits (e.g., percentages or sample counts for training, validation, or test sets) needed for reproduction. |
| Hardware Specification | No | The paper does not provide specific details about the hardware (e.g., GPU/CPU models, memory, or cloud instance types) used for running the experiments. |
| Software Dependencies | No | The paper does not specify any software dependencies with version numbers (e.g., programming languages, libraries, or frameworks) used in the experiments. |
| Experiment Setup | Yes | Here we describe the parameters and settings that were varied in the experiments that are presented. ... The best set of parameter values was selected for each algorithm and each target distribution, where the possible values were α {0.1η,0.5η,η,2η,4η,8η} and β {0.1,0.3,0.5,0.7,0.9}, with η > 0 problem-dependent (see the Supplement). |