TIGGER: Scalable Generative Modelling for Temporal Interaction Graphs

Authors: Shubham Gupta, Sahil Manchanda, Srikanta Bedathur, Sayan Ranu6819-6828

AAAI 2022 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Through extensive experiments on real datasets, we establish TIGGER generates graphs of superior fidelity, while also being up to 3 orders of magnitude faster than the state-of-the-art. In this section, we benchmark TIGGER against DYMOND and TAGGEN and establish that it (1) it is up to 2000 times faster, (2) breaks new ground on scalability against number of timestamps, and (3) generates graphs of high fidelity.
Researcher Affiliation Academia Department of Computer Science and Engineering Indian Institute of Technology, Delhi {shubham.gupta,sahil.manchanda,srikanta,sayanranu}@cse.iitd.ac.in
Pseudocode Yes Algorithm 1: Sampling synthetic temporal random walks from a trained transductive recurrent generative model; Algorithm 2: Sampling synthetic temporal random walks from a trained inductive recurrent generative model
Open Source Code Yes Our codebase and datasets are available at https://github.com/ data-iitd/tigger.
Open Datasets Yes Datasets: For our empirical evaluation, we use the publicly available datasets listed in Table 1. ... Our datasets span various domains including message exchange platform (UC Irvine) (Kunegis 2013a), financial network (Bitcoin) (Kumar et al. 2016), communication forum (Reddit) (Leskovec and Krevl 2014), shopping (Ta-feng) (Bai et al. 2018), and Wikipedia edits (Wiki) (Leskovec and Krevl 2014).
Dataset Splits No Parameter details along with machine configuration are provided in the supplementary material.
Hardware Specification No Parameter details along with machine configuration are provided in the supplementary material.
Software Dependencies No The paper does not provide specific version numbers for software dependencies or libraries used in the experiments.
Experiment Setup No Parameter details along with machine configuration are provided in the supplementary material.