Online Influence Maximization with Node-Level Feedback Using Standard Offline Oracles

Authors: Zhijie Zhang, Wei Chen, Xiaoming Sun, Jialin Zhang9153-9161

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

Reproducibility Variable Result LLM Response
Research Type Theoretical We resolve the aforementioned challenges for the IC model and present the first e O(poly(|G|) T)-regret algorithm with node-level feedback using standard offline oracles. In the technical part, our main contribution is a novel adaptation of the maximum likelihood estimation (MLE) approach which can learn the edge-level parameters and their confidence ellipsoids based on the node-level feedback. Further, we adjust the update procedure to dissect the confidence ellipsoid into confidence intervals on each parameter, so that we can apply a standard offline influence maximization oracle instead of the pair-oracle.
Researcher Affiliation Collaboration Zhijie Zhang,1,2 Wei Chen,3 Xiaoming Sun,1,2 Jialin Zhang1,2, 1 Institute of Computing Technology, Chinese Academy of Sciences, Beijing, China 2 University of Chinese Academy of Sciences, Beijing, China 3 Microsoft Research Asia, Beijing, China {zhangzhijie,sunxiaoming,zhangjialin}@ict.ac.cn, weic@microsoft.com
Pseudocode Yes Algorithm 1: IC-UCB and Algorithm 2: Estimate.
Open Source Code No The paper does not provide any statement or link indicating that source code for the described methodology is publicly available.
Open Datasets No The paper is theoretical and does not conduct experiments on a specific dataset. Therefore, it does not mention or provide access to any public datasets.
Dataset Splits No The paper is theoretical and does not conduct experiments with datasets, so it does not discuss training, validation, or test splits.
Hardware Specification No The paper is theoretical and does not report on experiments requiring specific hardware. Therefore, no hardware specifications are mentioned.
Software Dependencies No The paper is theoretical and does not report on experiments requiring specific software. Therefore, no software dependencies with version numbers are listed.
Experiment Setup No The paper is theoretical, presenting algorithms and their regret analysis, rather than empirical results from experiments. Therefore, no experimental setup details like hyperparameters or training settings are provided.