From which world is your graph

Authors: Cheng Li, Felix MF Wong, Zhenming Liu, Varun Kanade

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

Reproducibility Variable Result LLM Response
Research Type Experimental We report several experiments on synthetic and real-world data collected on Twitter from Oct 1 and Nov 30, 2016. Our experiments demonstrate that our model and inference algorithms perform well on real-world data and reveal interesting structures in networks.
Researcher Affiliation Collaboration Cheng Li College of William & Mary Felix M. F. Wong Independent Researcher Zhenming Liu College of William & Mary Varun Kanade University of Oxford Currently at Google.
Pseudocode Yes Figure 1: Subroutines of our Latent Inference Algorithm. LATENT-INFERENCE(A) 1 // Step 1. Estimate Φ . 2 bΦ = SM-EST(A). 3 // Step 2. Execute isomap algo. 4 D = ISOMAP-ALGO(bΦ) 5 // Step 3. Find latent variables. 6 Run a line embedding algorithm [48, 49]. ISOMAP-ALGO(bΦ, ℓ) 1 Execute S DENOISE(bΦ) (See Section 3.2) 2 // S is a subset of [n]. 3 Build G = {S, E} s.t. {i, j} E iff 4 |( Φd)i ( Φd)j| ℓ/ log n (ℓa constant). 5 Compute D such D(i, j) is the shortest 6 path distance between i and j when i, j S. 7 return D SM-EST(A, t) 1 [ UA, SA, VA] = svd(A). 2 Let also λi be i-th singular value of A. 3 // let t be a suitable parameter. 4 d = DECIDETHRESHOLD(t, ρ(n)). 5 SA: diagonal matrix comprised of {λi}i d 6 UA, VA: the singular vectors 7 corresponding to SA. 8 Let bΦ = p C(n)UAS1/2 A . 9 return bΦ DECIDETHRESHOLD(t, ρ(n)) 1 // This procedure decides d the number 2 of Eigenvectors to keep. 3 // t is a tunable parameter. See Proposition 3.1. 4 d = arg maxd{λd( A ρ(n)) λd+1( A ρ(n)) θ}. 5 where θ = 10(t/ρ(n))24/59
Open Source Code No The paper does not provide any concrete access (link, explicit statement of release) to the source code for the methodology described.
Open Datasets No The paper mentions 'real-world data collected on Twitter from Oct 1 and Nov 30, 2016' and 'Ideology scores of the US Congress (estimated by third parties [57])' as ground-truth. However, it does not provide specific links, DOIs, repositories, or explicit statements for public access to the collected Twitter dataset used in the experiments. While [57] is cited for ground truth, the primary dataset (Twitter data) is described as collected by the authors without public access details.
Dataset Splits No The paper does not provide specific details on dataset splits (e.g., percentages, sample counts, or references to predefined splits) for training, validation, or testing.
Hardware Specification No The paper does not explicitly describe the specific hardware (GPU, CPU models, memory, etc.) used to run the experiments.
Software Dependencies No The paper does not provide specific software dependencies with version numbers.
Experiment Setup No The paper does not provide specific experimental setup details such as hyperparameter values, optimizer settings, or other training configurations.