Learning Influence Functions from Incomplete Observations

Authors: Xinran He, Ke Xu, David Kempe, Yan Liu

NeurIPS 2016 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Experiments on synthetic and real-world datasets demonstrate the ability of our method to compensate even for a fairly large fraction of missing observations.
Researcher Affiliation Academia Xinran He Ke Xu David Kempe Yan Liu University of Southern California, Los Angeles, CA 90089 {xinranhe, xuk, dkempe, yanliu.cs}@usc.edu
Pseudocode No No structured pseudocode or algorithm blocks were found.
Open Source Code No The paper states: "We use the preprocessed version of the dataset released by Du et al. [3] and available at http://www.cc.gatech.edu/~ndu8/InfluLearner.html." This link is for a dataset and preprocessed version related to a baseline, not the authors' own methodology code. No other explicit statement or link for open-source code for their method was found.
Open Datasets Yes We further evaluate the performance of our method on the real-world Meme Tracker7 dataset [11]. The dataset consists of the propagation of short textual phrases... We use the preprocessed version of the dataset released by Du et al. [3] and available at http://www.cc.gatech.edu/~ndu8/InfluLearner.html.
Dataset Splits Yes Subsequently, we generate 8192 cascades as training data... The test set contains 200 independently sampled seed sets... We follow exactly the same evaluation method as Du et al. [3] with a training/test set split of 60%/40%.
Hardware Specification No No specific hardware details (e.g., GPU/CPU models, memory specifications) used for running experiments were mentioned in the paper.
Software Dependencies No No specific software dependencies with version numbers (e.g., Python 3.8, PyTorch 1.9) were mentioned in the paper.
Experiment Setup Yes For the model-free approaches (Influ Learner and our algorithm), we use K = 200 features.