Notice: The reproducibility variables underlying each score are classified using an automated LLM-based pipeline, validated against a manually labeled dataset. LLM-based classification introduces uncertainty; scores should be interpreted as estimates. Full accuracy metrics and methodology are described in [1]
Autoregressive Networks
Authors: Binyan Jiang, Jialiang Li, Qiwei Yao
JMLR 2023 | Venue PDF | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Simulation results are reported in Section 4, and the illustration with three real dynamic network data sets is presented in Section 5. |
| Researcher Affiliation | Academia | Binyan Jiang EMAIL Department of Applied Mathematics Hong Kong Polytechnic University Hong Kong, China Jialiang Li EMAIL Department of Statistics and Applied Probability National University of Singapore Singapore 117546 Qiwei Yao EMAIL Department of Statistics London School of Economics London, WC2A 2AE |
| Pseudocode | No | The paper describes a permutation algorithm with numbered steps in Section 2.4 (Model diagnostic check) but does not present it as a formally labeled 'Pseudocode' or 'Algorithm' block. Therefore, it does not meet the criteria for structured pseudocode. |
| Open Source Code | No | The paper does not contain any explicit statements or links indicating that the source code for the described methodology is publicly available. |
| Open Datasets | Yes | This data set is now available in R packages igraphdata and sand. ... The data are available at www.sociopatterns.org/datasets/ high-school-contact-and-friendship-networks/. ... The data used are a subset of the openly available trade data for 205 countries in 1870 2014 (Barbieri et al., 2009; Barbieri and Keshk, 2016). |
| Dataset Splits | No | The paper describes the processing of existing datasets (e.g., combining 24-hour periods for RFID data, removing individuals with missing values for French high school data, defining edges for global trade data), but it does not specify explicit training, validation, or test dataset splits for model evaluation or reproduction. |
| Hardware Specification | Yes | Using a computer with Intel(R) Core(TM) i7-10875H CPU and 32.0 GB RAM, we need to spend 0.36 and 252.39 seconds to obtain the community detection results with our method and dynsbm method, respectively. |
| Software Dependencies | No | To apply the variational EM algorithm of Matias and Miele (2017) to analyze this data set, we use the R package dynsbm. |
| Experiment Setup | Yes | We generate data according to model (1) in which the parameters αij and βij are drawn independently from U[0.1, 0.5]... We now consider model (14) with q = 2 or 3 clusters, in which θi,i = ηi,i = 0.4 for i = 1, , q, and θi,j and ηi,j, for 1 i, j q, are drawn independently from U[0.05, 0.25]... To choose the number of clusters q objectively, we define the Bayesian information criteria (BIC) as follows: BIC(q) = 2 max log(likelihood) + log{n(p/q)2}q(q + 1). |