Smoothing Structured Decomposable Circuits

Authors: Andy Shih, Guy Van den Broeck, Paul Beame, Antoine Amarilli

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

Reproducibility Variable Result LLM Response
Research Type Experimental Our final contribution (Section 8) is to experiment on smoothing and probabilistic inference tasks. We evaluate the performance of our smoothing and of our linear time All-Marginals algorithm.
Researcher Affiliation Academia Andy Shih University of California, Los Angeles andyshih@cs.ucla.edu Guy Van den Broeck University of California, Los Angeles guyvdb@cs.ucla.edu Paul Beame University of Washington beame@cs.washington.edu Antoine Amarilli LTCI, Télécom Paris, IP Paris antoine.amarilli@telecom-paris.fr
Pseudocode Yes Algorithm 1 all-marginals(g, w)
Open Source Code Yes The code for our experiments can be found at https://github.com/Andy Shih12/SSDC.
Open Datasets Yes In Table 2b we report the results on the Segmentation-11 network, which is a network from the 2006-2014 UAI Probabilistic Inference competitions. This particular network is a factor graph that was used to do image segmentation/classification (figure out what type of object each pixel corresponds to) [Forouzan, 2015].
Dataset Splits No The paper does not specify dataset splits such as train/validation/test percentages, absolute sample counts, or describe a cross-validation setup for its experiments.
Hardware Specification Yes Experiments were run on a single Intel(R) Core(TM) i7-3770 CPU with 16GB of RAM.
Software Dependencies No The paper does not provide specific version numbers for any software dependencies or libraries used in the experiments.
Experiment Setup No The paper describes the general setup of hand-crafted circuits and the context of collapsed sampling, but it does not provide specific experimental setup details such as hyperparameter values (e.g., learning rates, batch sizes) or optimizer settings.