Fair classification by loss balancing via fairness-aware batch sampling
Existing classification models often output discriminatory results since they learn the target
attribute without addressing data imbalance with respect to the protected attributes (eg …
attribute without addressing data imbalance with respect to the protected attributes (eg …
The Sharpe predictor for fairness in machine learning
S Liu, LN Vicente - arXiv preprint arXiv:2108.06415, 2021 - arxiv.org
In machine learning (ML) applications, unfair predictions may discriminate against a minority
group. Most existing approaches for fair machine learning (FML) treat fairness as a …
group. Most existing approaches for fair machine learning (FML) treat fairness as a …
Fair enough: Improving fairness in budget-constrained decision making using confidence thresholds
© 2020 for this paper by its authors. Increasing concern about discrimination and bias in
datadriven decision making systems has led to a growth in academic and popular interest in …
datadriven decision making systems has led to a growth in academic and popular interest in …
Fairness guarantees in multi-class classification with demographic parity
Algorithmic Fairness is an established area of machine learning, willing to reduce the
influence of hidden bias in the data. Yet, despite its wide range of applications, very few …
influence of hidden bias in the data. Yet, despite its wide range of applications, very few …
The impact of split classifiers on group fairness
Disparate treatment occurs when a machine learning model produces different decisions for
groups of individuals based on a sensitive attribute (eg, age, sex). In domains where …
groups of individuals based on a sensitive attribute (eg, age, sex). In domains where …
Addressing strategic manipulation disparities in fair classification
In real-world classification settings, such as loan application evaluation or content
moderation on online platforms, individuals respond to classifier predictions by strategically …
moderation on online platforms, individuals respond to classifier predictions by strategically …
On the power of randomization in fair classification and representation
S Agarwal, A Deshpande - Proceedings of the 2022 ACM Conference …, 2022 - dl.acm.org
Fair classification and fair representation learning are two important problems in supervised
and unsupervised fair machine learning, respectively. Fair classification asks for a classifier …
and unsupervised fair machine learning, respectively. Fair classification asks for a classifier …
Minimax pareto fairness: A multi objective perspective
In this work we formulate and formally characterize group fairness as a multi-objective
optimization problem, where each sensitive group risk is a separate objective. We propose a …
optimization problem, where each sensitive group risk is a separate objective. We propose a …
Forml: Learning to reweight data for fairness
B Yan, S Seto, N Apostoloff - arXiv preprint arXiv:2202.01719, 2022 - arxiv.org
Machine learning models are trained to minimize the mean loss for a single metric, and thus
typically do not consider fairness and robustness. Neglecting such metrics in training can …
typically do not consider fairness and robustness. Neglecting such metrics in training can …
Fairness with minimal harm: A pareto-optimal approach for healthcare
Common fairness definitions in machine learning focus on balancing notions of disparity
and utility. In this work, we study fairness in the context of risk disparity among sub …
and utility. In this work, we study fairness in the context of risk disparity among sub …