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].
FNNC: Achieving Fairness through Neural Networks
Authors: Manisha Padala, Sujit Gujar
IJCAI 2020 | Venue PDF | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Our experiments show that FNNC performs as good as the state of the art, if not better. The experimental evidence supplements our theoretical guarantees. |
| Researcher Affiliation | Academia | Manisha Padala and Sujit Gujar International Institute of Information and Technology, Hyderabad EMAIL, EMAIL |
| Pseudocode | No | The paper does not contain structured pseudocode or algorithm blocks. |
| Open Source Code | No | The paper does not provide an explicit statement or a link indicating the availability of open-source code for the described methodology. |
| Open Datasets | Yes | We have conducted experiments on the six most common datasets used in fairness domain. In Adult, Default, and German dataset... In Default and Compass datasets that we used... In the Bank dataset... |
| Dataset Splits | Yes | The results are averaged over 5-fold cross-validation performance on the data. |
| Hardware Specification | No | The paper does not provide specific details about the hardware used for running experiments, such as GPU or CPU models. |
| Software Dependencies | No | The paper does not specify software dependencies with version numbers, such as programming language versions or library versions. |
| Experiment Setup | Yes | The number of hidden neurons in both the layers was one of the following (100, 50), (200, 100), (500, 100). ... We fix the batch size to be either 1000 or 500 depending on the dataset... For training, we have used the Adam Optimizer with a learning rate of 0.01 or 0.001 and the training typically continues for a maximum of 5000 epochs for each experiment before convergence. |