Expectation Particle Belief Propagation
Authors: Thibaut Lienart, Yee Whye Teh, Arnaud Doucet
NeurIPS 2015 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We investigate the performance of our method on MRFs for two simple graphs. This allows us to compare the performance of EPBP to the performance of PBP in depth. We also illustrate the behavior of the sub-quadratic version of EPBP. Finally we show that EPBP provides good results in a simple denoising application. |
| Researcher Affiliation | Academia | Thibaut Lienart, Yee Whye Teh, Arnaud Doucet Department of Statistics University of Oxford Oxford, UK {lienart,teh,doucet}@stats.ox.ac.uk |
| Pseudocode | Yes | Algorithm 1 Node update |
| Open Source Code | Yes | All simulations were run with Julia on a Mac with 2.5 GHz Intel Core i5 processor, our code is available online.2 (2https://github.com/tlienart/EPBP.) |
| Open Datasets | No | The paper describes generating data based on specific potentials (e.g., equations 14 and 15) for a 3x3 grid and a tree, but no concrete access information (link, DOI, specific repository, or formal citation for a public dataset) is provided for the datasets used in the experiments. |
| Dataset Splits | No | The paper does not provide specific details on dataset splits (e.g., percentages, sample counts, or citations to predefined splits) for training, validation, or testing. |
| Hardware Specification | Yes | All simulations were run with Julia on a Mac with 2.5 GHz Intel Core i5 processor |
| Software Dependencies | No | The paper mentions 'Julia' as the programming language, but does not provide specific version numbers for any ancillary software dependencies, libraries, or solvers used in the experiments. |
| Experiment Setup | Yes | We compare the two methods after 20 LBP iterations. The results are obtained with N = 100 samples on each node and 20 BP iterations. The image has size 50 x 50 and the simulation was run with N = 30 particles per nodes, M = 5 and 10 BP iterations taking under 2 minutes to complete. |