Graph Estimation From Multi-Attribute Data

Authors: Mladen Kolar, Han Liu, Eric P. Xing

JMLR 2014 | Venue PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Extensive simulation studies demonstrate the effectiveness of the method under various conditions. We provide illustrative applications to uncovering gene regulatory networks from gene and protein profiles, and uncovering brain connectivity graph from positron emission tomography data.
Researcher Affiliation Academia Mladen Kolar EMAIL The University of Chicago Booth School of Business Chicago, Illinois 0637, USA; Han Liu EMAIL Department of Operations Research and Financial Engineering Princeton University Princeton, New Jersey 08544, USA; Eric P. Xing EMAIL Machine Learning Department Carnegie Mellon University Pittsburgh, Pennsylvania 15213, USA
Pseudocode Yes 3. Repeat Step 2 until the duality gap |||trp SpΩq log |pΩ| λ ∑a,b ||pΩab||F − ∑j PV kj log |Σ|||| ď ϵ, where ϵ is a prefixed precision parameter (for example, ϵ 10−3).
Open Source Code No The paper does not provide explicit access to source code for the methodology described. It mentions other software packages like 'glasso and huge' but not their own implementation.
Open Datasets Yes We apply our method to the Positron Emission Tomography data set, which contains 259 subjects... The data can be obtained from http://adni.loni.ucla.edu/.
Dataset Splits Yes Next, we use the selected λ to estimate 100 networks based on random subsamples containing 80% of the data-points.
Hardware Specification No The paper mentions 'resources provided by the University of Chicago Research Computing Center' in the acknowledgments, which is a general computing environment, but does not provide specific hardware details (e.g., CPU/GPU models, memory).
Software Dependencies No The paper does not provide specific software names with version numbers for the implementation of their method. It mentions 'glasso and huge' as existing packages but without version details relevant to their work.
Experiment Setup No The paper discusses selecting tuning parameters by minimizing the Bayesian information criterion 'over a grid of parameter values' and using cross-validation, but it does not specify concrete values for hyperparameters like the ranges of the grid or other system-level training settings.