Enhancing Model Fairness and Accuracy with Similarity Networks: A Methodological Approach
This addresses fairness issues in ML models for applications requiring unbiased predictions, though it appears incremental as an extension of similarity-based methods.
The paper tackles bias in machine learning datasets by mapping instances into a similarity feature space with adjustable resolution, showing that this approach improves model fairness and accuracy in classification, imputation, and augmentation tasks while satisfying criteria like demographic parity.
In this paper, we propose an innovative approach to thoroughly explore dataset features that introduce bias in downstream machine-learning tasks. Depending on the data format, we use different techniques to map instances into a similarity feature space. Our method's ability to adjust the resolution of pairwise similarity provides clear insights into the relationship between the dataset classification complexity and model fairness. Experimental results confirm the promising applicability of the similarity network in promoting fair models. Moreover, leveraging our methodology not only seems promising in providing a fair downstream task such as classification, it also performs well in imputation and augmentation of the dataset satisfying the fairness criteria such as demographic parity and imbalanced classes.