On clustering network-valued data

Authors: Soumendu Sundar Mukherjee, Purnamrita Sarkar, Lizhen Lin

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

Reproducibility Variable Result LLM Response
Research Type Experimental We illustrate our methods using both simulated and real data sets, and theoretical justifications are provided in terms of consistency.
Researcher Affiliation Academia Soumendu Sundar Mukherjee Department of Statistics University of California, Berkeley Berkeley, California 94720, USA soumendu@berkeley.edu Purnamrita Sarkar Department of Statistics and Data Sciences University of Texas, Austin Austin, Texas 78712, USA purna.sarkar@austin.utexas.edu Lizhen Lin Department of Applied and Computational Mathematics and Statistics Univeristy of Notre Dame Notre Dame, Indiana 46556, USA lizhen.lin@nd.edu
Pseudocode Yes Algorithm 1 Network Clustering based on Graphon Estimates (NCGE) Algorithm 2 Network Clustering based on Log Moments (NCLM)
Open Source Code Yes Code used in this paper is publicly available at https://github.com/soumendu041/clustering-network-valued-data.
Open Datasets Yes We cluster about fifty real world networks. We use 11 co-authorship networks between 15,000 researchers from the High Energy Physics corpus of the ar Xiv, 11 co-authorship networks with 21,000 nodes from Citeseer (which had Machine Learning in their abstracts), 17 co-authorship networks (each with about 3000 nodes) from the NIPS conference and finally 10 Facebook ego networks2. ... 2https://snap.stanford.edu/data/egonets-Facebook.html
Dataset Splits No The paper mentions using simulated and real data for experiments but does not provide specific train/validation/test splits or cross-validation details for reproduction.
Hardware Specification No The paper does not provide specific hardware details (e.g., CPU/GPU models, memory) used for running the experiments.
Software Dependencies No The paper does not list specific software dependencies with version numbers required to reproduce the experiments.
Experiment Setup No The paper describes the algorithms and their theoretical properties but does not provide concrete hyperparameter values or system-level training settings for the experiments.