Notice: The reproducibility variables underlying each score are classified using an automated LLM-based pipeline, validated against a manually labeled dataset. LLM-based classification introduces uncertainty and potential bias; scores should be interpreted as estimates. Full accuracy metrics and methodology are described in [1].
Fast&Fair: Training Acceleration and Bias Mitigation for GNNs
Authors: Oyku Deniz Kose, Yanning Shen
TMLR 2023 | Venue PDF | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Experiments on node classification over real-world networks demonstrate the efficiency of the proposed scheme in improving fairness in terms of statistical parity and equal opportunity compared to fairness-aware baselines. In addition, it is empirically shown that the proposed framework leads to faster convergence compared to the naive baseline where no normalization is employed. (Abstract) and In this section, experimental results obtained on real-world datasets for a supervised node classification task are presented. The performance of the proposed framework, Fair Norm, is compared with baseline schemes in terms of node classification accuracy and fairness metrics. Furthermore, the influence of the proposed fairness-aware normalization strategy on convergence speed is examined. (Section 5, Experiments) |
| Researcher Affiliation | Academia | O. Deniz Kose EMAIL Department of Electrical Engineering and Computer Science University of California Irvine Yanning Shen EMAIL Department of Electrical Engineering and Computer Science University of California Irvine |
| Pseudocode | No | The paper describes methods through mathematical formulations, theorems, and proofs but does not contain explicitly labeled pseudocode or algorithm blocks. |
| Open Source Code | No | The paper does not contain an explicit statement or link indicating the release of source code for the described methodology. |
| Open Datasets | Yes | In the experiments, three real-world networks are used: Pokec-z, Pokec-n (Dai & Wang, 2021), and the Recidivism graph (Jordan & Freiburger, 2015). Pokec-z and Pokec-n are created by sampling the anonymized, 2012 version of Pokec (Takac & Zabovsky, 2012), which is a social network used in Slovakia (Dai & Wang, 2021). The information of defendants (corresponding to nodes) who got released on bail at the U.S. state courts during 1990-2009 (Jordan & Freiburger, 2015) is utilized to build the Recidivism graph... |
| Dataset Splits | Yes | Furthermore, training of the model is executed over 50% of the nodes, while the remaining nodes are equally divided to be used as the validation and test sets. For each experiment, results for five random data splits are obtained, and the average of them together with standard deviations are presented. |
| Hardware Specification | No | The paper does not provide specific hardware details (e.g., GPU/CPU models, processor types) used for running the experiments. |
| Software Dependencies | No | The paper mentions the use of Adam optimizer and GCN, but does not specify any particular software libraries (e.g., PyTorch, TensorFlow) or their version numbers. |
| Experiment Setup | Yes | A two-layer GCN (Kipf & Welling, 2017) followed by a linear layer is employed for the classification task... All models are trained for 1000 epochs by employing Adam optimizer (Kingma & Ba, 2014) together with a learning rate of 10-3 and ℓ2 weight decay factor of 10-5. Hidden dimension of the node representations is selected as 64 on all datasets. The hyperparameters of the proposed fairness-aware framework and all other baselines are tuned via a grid search on cross-validation sets, see again Appendix D for the utilized hyperparameter values. |