Fair Attribute Completion on Graph with Missing Attributes

Authors: Dongliang Guo, Zhixuan Chu, Sheng Li

ICLR 2023 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Experimental results on benchmark datasets show that our method achieves better fairness performance with less sacrifice in accuracy, compared with the state-of-the-art methods of fair graph learning.
Researcher Affiliation Collaboration Dongliang Guo1, Zhixuan Chu2, Sheng Li3,1 1University of Georgia, 2Ant Group, 3University of Virginia
Pseudocode Yes Algorithm 1 Fair AC framework algorithm
Open Source Code Yes Code is available at: https://github.com/donglgcn/Fair AC.
Open Datasets Yes In the experiments, we use three public graph datasets, NBA, Pokec-z, and Pokec-n. ... The NBA dataset (Dai & Wang, 2021)... Pokec (Takac & Zabovsky, 2012)...
Dataset Splits No The paper specifies a 'training/test set' split ('75%/25%') but does not explicitly mention a separate 'validation' split or its percentage.
Hardware Specification No The paper does not provide specific hardware details such as CPU models, GPU models, or memory specifications used for running experiments.
Software Dependencies No The paper mentions using 'Adam' optimizer and the 'Karate Club library' for Deep Walk implementation, but it does not specify version numbers for these or other software dependencies like Python, PyTorch, or CUDA.
Experiment Setup Yes We adopt Adam (Kingma & Ba, 2014) with the learning rate of 0.001 and weight decay as 1e 5. ... We set walk length as 100, embedding dimension as 64, window size as 5, and epochs as 10.