On Dyadic Fairness: Exploring and Mitigating Bias in Graph Connections

Authors: Peizhao Li, Yifei Wang, Han Zhao, Pengyu Hong, Hongfu Liu

ICLR 2021 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Empirical validation demonstrates that our method delivers effective dyadic fairness in terms of various statistics, and at the same time enjoys a favorable fairness-utility tradeoff.
Researcher Affiliation Academia 1Brandeis University, 2University of Illinois at Urbana-Champaign {peizhaoli,yifeiwang,hongpeng,hongfuliu}@brandeis.edu hanzhao@illinois.edu
Pseudocode Yes Algorithm 1: Algorithmic routine for Fair Adj
Open Source Code No The paper does not provide an explicit statement or link to its open-source code.
Open Datasets Yes We conduct experiments on real-world social networks and citation networks including Oklahoma97, UNC28 (Traud et al., 2011), Facebook#1684, Cora, Citeseer, and Pubmed.
Dataset Splits No The paper mentions "train/test splits" but does not explicitly detail a separate validation split or its proportion/use.
Hardware Specification Yes Experiments are conducted on Nvidia Titan RTX graphics card.
Software Dependencies No The paper mentions various methods (e.g., VGAE, node2vec, Fairwalk) but does not specify software or library version numbers for reproducibility.
Experiment Setup Yes For all experiments, we set T1 = 50 and the total epochs which contain T1 and T2 equal to 4. Graph neural networks are applied with two hidden layers with size 32 and 16 respectively. ηθ is set to 0.01. For η e A for different datasets, we have: Oklahoma97: 0.1; UNC28: 0.1; Cora: 0.2; Citeseer: 0.5.