Faster Fair Machine via Transferring Fairness Constraints to Virtual Samples
Authors: Zhou Zhai, Lei Luo, Heng Huang, Bin Gu
AAAI 2023 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Empirically, we test the proposed method on real-world datasets and all results confirm its excellent performance. |
| Researcher Affiliation | Academia | 1 School of Computer and Software, Nanjing University of Information Science and Technology, P.R.China 2 School of Computer Science and Engineering, Nanjing University of Science and Technology, P.R.China 3 Department of Electrical & Computer Engineering, University of Pittsburgh, USA 4 Department of Machine Learning, MBZUAI, United Arab Emirates |
| Pseudocode | No | The paper describes the method using mathematical formulations and descriptive text, but does not include any explicitly labeled 'Pseudocode' or 'Algorithm' blocks. |
| Open Source Code | No | The paper does not provide any specific links to source code repositories, nor does it state that the code will be released or is available in supplementary materials. |
| Open Datasets | Yes | Datesets: Table 2 summarizes the datasets used in the experiments. We use two synthetic datasets (i.e., Synth Opp and Synth Same) whose details are provided in Appendix. The sensitive attribute is listed in the column Sensitive Attribute of Table 2. |
| Dataset Splits | No | The paper states 'We repeatedly split each dataset into a train (75%) and test (25%) set.' and does not mention a distinct validation split. |
| Hardware Specification | No | The paper does not provide any specific details regarding the hardware (e.g., GPU models, CPU types, or cloud configurations) used to conduct the experiments. |
| Software Dependencies | No | The paper mentions 'LIBSVM (Chang and Lin 2011)' and 'the quadprog function in MATLAB' but does not provide specific version numbers for these or any other software dependencies. |
| Experiment Setup | Yes | The regularization parameter C in SVM is fixed at 10. The Gaussian kernel K(x1, x2) = exp(κ x1 x2 2) with κ = 0.5 and Polynomial kernel K(x1, x2) = (x1x2 + 1)2 is used in all the experiments. For each dataset, we first calculate the fairness score c = 1/n Pn i=1(zi z)g(yi, xi) of the model without fairness constraints. Then, we set the fairness constrain value c = 1/2c for our proposed methods. The compared methods are the same setting. |