# Using Pareto Simulated Annealing to Address Algorithmic Bias in Machine Learning

arXiv: https://arxiv.org/abs/2105.15064 · HTML: https://mlsanity.com/paper/2105.15064/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Builds on 9 of its 31 references that are in the map.
- First posted on arXiv 31 May 2021.

## Builds on (most cited first, 9 of 9)

- [Certifying and removing disparate impact](https://mlsanity.com/paper/1412.3756/) (2014)
- [Fairness in Machine Learning: A Survey](https://mlsanity.com/paper/2010.04053/) (2020)
- [Fairness Beyond Disparate Treatment & Disparate Impact: Learning Classification without Disparate Mistreatment](https://mlsanity.com/paper/1610.08452/) (2016)
- [Characterising Bias in Compressed Models](https://mlsanity.com/paper/2010.03058/) (2020)
- [Satisfying Real-world Goals with Dataset Constraints](https://mlsanity.com/paper/1606.07558/) (2016)
- [Algorithmic Factors Influencing Bias in Machine Learning](https://mlsanity.com/paper/2104.14014/) (2021)
- [Fairness-aware machine learning: a perspective](https://mlsanity.com/paper/1708.00754/) (2017)
- [Underestimation Bias and Underfitting in Machine Learning](https://mlsanity.com/paper/2005.09052/) (2020)
- [Customizing Pareto Simulated Annealing for Multi-objective Optimization of Control Cabinet Layout](https://mlsanity.com/paper/1906.04825/) (2019)

## Built on by (most cited first, 3 of 3)

- [Bias Mitigation for Machine Learning Classifiers: A Comprehensive Survey](https://mlsanity.com/paper/2207.07068/) (2022)
- [Introducing a Family of Synthetic Datasets for Research on Bias in Machine Learning](https://mlsanity.com/paper/2107.08928/) (2021)
- [Genetic Generalized Additive Models](https://mlsanity.com/paper/2602.15877/) (2026)

## Abstract (arXiv)

Algorithmic Bias can be due to bias in the training data or issues with the algorithm itself. These algorithmic issues typically relate to problems with model capacity and regularisation. This underestimation bias may arise because the model has been optimised for good generalisation accuracy without any explicit consideration of bias or fairness. In a sense, we should not be surprised that a model might be biased when it hasn't been "asked" not to be. In this paper, we consider including bias (underestimation) as an additional criterion in model training. We present a multi-objective optimisation strategy using Pareto Simulated Annealing that optimise for both balanced accuracy and underestimation. We demonstrate the effectiveness of this strategy on one synthetic and two real-world datasets.