Continuous Conditional Dependency Network for Structured Regression

Authors: Chao Han, Mohamed Ghalwash, Zoran Obradovic

AAAI 2017 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental The performance of CCDN on multiple datasets provides evidence of its structure recovery ability and superior effectiveness and efficiency as compared to the state-of-the-art alternatives.
Researcher Affiliation Collaboration Chao Han Center for Data Analytics and Biomedical Informatics, Temple University Philadelphia, PA 19122 USA chao.han@temple.edu Mohamed Ghalwash Center for Computational Health IBM T.J. Watson Research Center Cambridge, MA, USA Temple University, PA, USA Ain Shams University, Egypt, Cairo mohamed.ghalwash@temple.edu Zoran Obradovic Center for Data Analytics and Biomedical Informatics, Temple University Philadelphia, PA 19122 USA zoran.obradovic@temple.edu
Pseudocode Yes Algorithm 1 Iterated Conditional Mode
Open Source Code No The paper does not include an unambiguous statement that the authors are releasing the code for the work described in this paper, nor does it provide a direct link to a source-code repository.
Open Datasets Yes Wind power data is obtained from the Global Energy Forecasting 2012 competition1. Precipitation Forecasting. It can be downloaded from NOAA s NCDC website2. Solar Energy Forecasting. The data is available on kaggle 3.
Dataset Splits Yes We considered 4 different training sizes r = {60, 120, 180, 240}, and fix validation and testing sizes as 60.
Hardware Specification No The paper states 'The experiments on different data were running on different nodes of cluster.' but provides no specific hardware details such as CPU/GPU models or memory specifications.
Software Dependencies No The paper does not provide specific software dependencies, such as library names with version numbers, needed to replicate the experiment.
Experiment Setup Yes The λ of SGCRF is picked from {1e 4, 1e 3, 1e 2, 0.1, 1} via 8 folds cross-validation. The neural network is set with one hidden layer with 20 hidden neurons.