Tag Archives: Fernandes Machado

EquiPy: Sequential Fairness using Optimal Transport in Python

Our article EquiPy: Sequential Fairness using Optimal Transport in Python, with Agathe Fernandes Machado, Suzie Grondin, François Hu and Philipp Ratz is now online. See also equilibration.github.io/equipy/ for the Python package

Algorithmic fairness has received considerable attention due to the failures of various predictive AI systems that have been found to be unfairly biased against subgroups of the population. Many approaches have been proposed to mitigate such biases in predictive systems, however, they often struggle to provide accurate estimates and transparent correction mechanisms in the case where multiple sensitive variables, such as a combination of gender and race, are involved. This paper introduces a new open source Python package, EquiPy, which provides a easy-to-use and model agnostic toolbox for efficiently achieving fairness across multiple sensitive variables. It also offers comprehensive graphic utilities to enable the user to interpret the influence of each sensitive variable within a global context. EquiPy makes use of theoretical results that allow the complexity arising from the use of multiple variables to be broken down into easier-to-solve sub-problems. We demonstrate the ease of use for both mitigation and interpretation on publicly available data derived from the US Census and provide sample code for its use.

Post-Calibration Techniques: Balancing Calibration and Score Distribution Alignment (NeurIPS’24)

Agathe Fernandes Machado will soon be on her way to Vancouver. She will attend the Thirty-Eighth Annual Conference on Neural Information Processing Systems (also known as NeurIPS 2024), to present a short paper on Post-Calibration Techniques: Balancing Calibration and Score Distribution Alignment

A binary scoring classifier can appear well-calibrated according to standard calibration metrics, even when the distribution of scores does not align with the distribution of the true events. In this paper, we investigate the impact of post-processing calibration on the score distribution (sometimes named “recalibration”). Using simulated data, where the true probability is known, followed by real-world datasets with prior knowledge on event distributions, we compare the performance of an XGBoost model before and after applying calibration techniques. The results show that while applying methods such as Platt scaling, Beta calibration, or isotonic regression can improve the model’s calibration, they may also lead to an increase in the divergence between the score distribution and the underlying event probability distribution.

Probabilistic Scores of Classifiers, Calibration is not Enough

Our paper “Probabilistic Scores of Classifiers, Calibration is not Enough”, with Agathe Fernandes Machado, Emmanuel Flachaire, Ewen Gallic and François Hu is now available on https://arxiv.org/abs/2408.03421

In binary classification tasks, accurate representation of probabilistic predictions is essential for various real-world applications such as predicting payment defaults or assessing medical risks. The model must then be well-calibrated to ensure alignment between predicted probabilities and actual outcomes. However, when score heterogeneity deviates from the underlying data probability distribution, traditional calibration metrics lose reliability, failing to align score distribution with actual probabilities. In this study, we highlight approaches that prioritize optimizing the alignment between predicted scores and true probability distributions over minimizing traditional performance or calibration metrics. When employing tree-based models such as Random Forest and XGBoost, our analysis emphasizes the flexibility these models offer in tuning hyperparameters to minimize the Kullback-Leibler (KL) divergence between predicted and true distributions. Through extensive empirical analysis across 10 UCI datasets and simulations, we demonstrate that optimizing tree-based models based on KL divergence yields superior alignment between predicted scores and actual probabilities without significant performance loss. In real-world scenarios, the reference probability is determined a priori as a Beta distribution estimated through maximum likelihood. Conversely, minimizing traditional calibration metrics may lead to suboptimal results, characterized by notable performance declines and inferior KL values. Our findings reveal limitations in traditional calibration metrics, which could undermine the reliability of predictive models for critical decision-making.

Sequential Conditional Transport on Probabilistic Graphs for Interpretable Counterfactual Fairness

Our paper “Sequential Conditional Transport on Probabilistic Graphs for Interpretable Counterfactual Fairness“, written with Agathe Fernandes Machado and Ewen Gallic, is now online

In this paper, we link two existing approaches to derive counterfactuals: adaptations based on a causal graph, as suggested in Plečko and Meinshausen (2020) and optimal transport, as in De Lara et al. (2024). We extend “Knothe’s rearrangement” Bonnotte (2013) and “triangular transport” Zech and Marzouk (2022) to probabilistic graphical models, and use this counterfactual approach, referred to as sequential transport, to discuss individual fairness. After establishing the theoretical foundations of the proposed method, we demonstrate its application through numerical experiments on both synthetic and real datasets.

Geospatial Disparities: A Case Study on Real Estate Prices in Paris

Our paper, Geospatial Disparities: A Case Study on Real Estate Prices in Paris, and Agathe Fernandes Machado, François Hu, Philipp Ratz and Ewen Gallic, is now online on ArXiv,

Driven by an increasing prevalence of trackers, ever more IoT sensors, and the declining cost of computing power, geospatial information has come to play a pivotal role in contemporary predictive models. While enhancing prognostic performance, geospatial data also has the potential to perpetuate many historical socio-economic patterns, raising concerns about a resurgence of biases and exclusionary practices, with their disproportionate impacts on society. Addressing this, our paper emphasizes the crucial need to identify and rectify such biases and calibration errors in predictive models, particularly as algorithms become more intricate and less interpretable. The increasing granularity of geospatial information further introduces ethical concerns, as choosing different geographical scales may exacerbate disparities akin to redlining and exclusionary zoning. To address these issues, we propose a toolkit for identifying and mitigating biases arising from geospatial data. Extending classical fairness definitions, we incorporate an ordinal regression case with spatial attributes, deviating from the binary classification focus. This extension allows us to gauge disparities stemming from data aggregation levels and advocates for a less interfering correction approach. Illustrating our methodology using a Parisian real estate dataset, we showcase practical applications and scrutinize the implications of choosing geographical aggregation levels for fairness and calibration measures.

The whole is greater than the sum of the parts

Good news: ou paper, A Sequentially Fair Mechanism for Multiple Sensitive Attributes, written with Philipp Ratz and François Hu will be presented in February in Vancouver, at the 38th Annual AAAI Conference on Artificial Intelligence. For a shorter version, there was a review on montrealethics.ai of the paper last week (as mentioned previously).

Also, last week also, the team launched the  equipy python package, with codes used in the paper,

pip install equipy

EquiPy is a Python package implementing sequential fairness on the predicted outputs of Machine Learning models, when dealing with multiple sensitive attributes. This post-processing method progressively achieve fairness accross a set of sensitive features by leveraging multi-marginal Wasserstein barycenters, which extends the standard notion of Strong Demographic Parity to the case with multiple sensitive characteristics. This approach seamlessly extends to approximate fairness, enveloping a framework accommodating the trade-off between performance and unfairness.

(from the left to the right, Agathe, who just joint the PhD program, Suzie, MSc student at ENSAE, with us since May or June, Philipp, PhD student, François, postdoctoral fellow – and Dante, also postdoctoral fellow, in stochastic processes). According to Aristotle (or probably slightly misquoted),

the whole is greater than the sum of the parts

I couldn’t agree more !