Learning from Contagion (Without Timestamps)

Authors: Kareem Amin, Hoda Heidari, Michael Kearns

ICML 2014 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We provide empirical evidence that our algorithm performs well more generally on realistic sparse graphs. 5.1. Experiments
Researcher Affiliation Academia Computer and Information Science, University of Pennsylvania
Pseudocode Yes Algorithm 1 Algorithm for exactly learning trees under passive seeds and independent cascade. Algorithm 2 K-lifts algorithm.
Open Source Code No The paper does not provide any specific links or explicit statements about the availability of source code for the described methodology.
Open Datasets Yes On the real collaboration network, Net Science (Newman, 2003; Boccaletti et al., 2006; Newman, 2006)
Dataset Splits No The paper does not provide specific details on train/validation/test dataset splits.
Hardware Specification No No specific hardware details (like GPU/CPU models or cloud instance types) used for running experiments were mentioned in the paper.
Software Dependencies No No specific software dependencies with version numbers were mentioned in the paper.
Experiment Setup Yes In all the experiments, each vertex becomes initially infected independently and with probability 0.05 (so roughly 5 vertices are seeded.) For simplicity, we will set the algorithm s parameter K = m(G), the correct number of edges, in each experiment. We run the experiments for M {1000, 10000, 100000, 1000000} and observe the relation between the performance of the K-lifts algorithm and the number of samples.