Differentially Private Learning of Undirected Graphical Models Using Collective Graphical Models

Authors: Garrett Bernstein, Ryan McKenna, Tao Sun, Daniel Sheldon, Michael Hay, Gerome Miklau

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

Reproducibility Variable Result LLM Response
Research Type Experimental We conduct a number of experiments on synthetic and real data to evaluate the quality of models learned by both Naive MLE and CGM.
Researcher Affiliation Academia 1University of Massachusetts Amherst 2Mount Holyoke College 3Colgate University. Correspondence to: Garrett Bernstein <gbernstein@cs.umass.edu>.
Pseudocode Yes Algorithm 1 Non-Linear Belief Propagation (NLBP) ... Algorithm 2 EM for CGMs
Open Source Code No No statement regarding the release or availability of open-source code for the described methodology was found.
Open Datasets No The paper uses synthetic data and human mobility data, but provides no concrete access information (links, DOIs, repository names, or citations with author/year) for either dataset.
Dataset Splits No The paper mentions reserving 25% of individuals' data for testing but does not specify training/validation/test splits with exact percentages, sample counts, or a clear splitting methodology for reproduction.
Hardware Specification No No specific hardware details (e.g., CPU/GPU models, memory, or cloud instance types) used for running experiments are mentioned in the paper.
Software Dependencies No No specific software dependencies with version numbers (e.g., Python 3.8, PyTorch 1.9) are provided.
Experiment Setup No The paper mentions data preprocessing steps and that PSGD was tuned via grid search, but it does not provide specific hyperparameter values, training configurations, or system-level settings for the models.