Study shows explanation disparities in machine learning models are influenced by data and model properties.
problem Disparities in post-hoc machine learning explanation methods across race and gender.
method Simulations and experiments on a real-world dataset to assess challenges to explanation disparities.
result Increased covariate shift, concept shift, and omission of covariates increase explanation disparities, especially for neural network models.
Recent work has shown that optical flow estimation can be formulated as a supervised learning task and can be successfully solved with convolutional networks. Training of the so-called FlowNet was enabled by a large synthetically generated dataset. The present paper extends the concept of optical flow estimation via co…
Paper explores fair classification with bounded disparity using finite datasets.
problem Ensuring fairness in binary classification with protected groups.
method Minimax optimal approach with fairness constraints and demographic disparity control.
result Proposes FairBayes-DDP+ method that achieves minimax lower bound on fairness-aware excess risk.
Deep chest X-ray classifiers show bias in predicting diagnoses.
problem Bias in deep learning classifiers predicting diagnoses from chest X-rays.
method Trained convolutional neural networks on multiple public datasets to predict 14 diagnostic labels.
result True positive rates vary significantly among different protected attributes, indicating bias.
The paper studies and mitigates accuracy disparity in regression models.
problem Accuracy disparity between different demographic subgroups in high-stakes domains.
method Error decomposition theorem and distribution alignment algorithm.
result The proposed algorithm effectively mitigates accuracy disparity while maintaining predictive power.
Study reveals class disparities in balanced datasets through spectral imbalance.
problem Class disparities in balanced datasets are overlooked despite model performance gaps.
method Developed a theoretical framework and studied 11 encoders to diagnose spectral imbalance.
result Identified spectral imbalance as a source of class disparities in balanced datasets.
Bayesian model identifies health disparities in disease progression.
problem Health disparities bias disease progression models.
method Interpretable Bayesian model accounting for three disparities.
result Model identifies and corrects for health disparities.
Following related work in law and policy, two notions of disparity have come to shape the study of fairness in algorithmic decision-making. Algorithms exhibit treatment disparity if they formally treat members of protected subgroups differently; algorithms exhibit impact disparity when outcomes differ across subgroups,…
End-to-end framework learns precise disparity for activity recognition.
problem Precise portrayal of intraclass disparity in activity recognition.
method Knowledge-directed adversarial learning framework with two competitive encoding distributions.
result Robust and generalizable performance on HAR benchmark datasets.
The paper examines how adversarial robustness affects accuracy disparity across different classes.
problem Understanding the impact of adversarial robustness on accuracy disparity across different classes.
method Linear classifiers under a Gaussian mixture model, decomposing the impact into inherent and imbalance effects.
result Adversarial robustness consistently degrades standard accuracy in balanced classes, but the class imbalance ratio plays a different role in accuracy disparity.
Study shows label errors impact model disparity metrics, proposing mitigation methods.
problem Impact of label errors on model disparity metrics.
method Empirical study, characterizing label error effects; proposing estimation and relabeling methods.
result Label errors significantly affect model disparity metrics, particularly for minority groups.
We combine multi-task learning and semi-supervised learning by inducing a joint embedding space between disparate label spaces and learning transfer functions between label embeddings, enabling us to jointly leverage unlabelled data and auxiliary, annotated datasets. We evaluate our approach on a variety of sequence cl…
Paper tackles fairness in CCA by minimizing correlation disparity error.
problem Fairness issues in CCA.
method Framework to minimize correlation disparity error in CCA.
result Reduces correlation disparity error without sacrificing CCA accuracy.
Automated data-driven decision making systems are increasingly being used to assist, or even replace humans in many settings. These systems function by learning from historical decisions, often taken by humans. In order to maximize the utility of these systems (or, classifiers), their training involves minimizing the e…
When the performance of a machine learning model varies over groups defined by sensitive attributes (e.g., gender or ethnicity), the performance disparity can be expressed in terms of the probability distributions of the input and output variables over each group. In this paper, we exploit this fact to reduce the dispa…
FairWASP optimizes training data to reduce disparities across subgroups.
problem Reducing disparities in model outputs across different subgroups in machine learning.
method A novel pre-processing approach that minimizes Wasserstein distance to the original dataset while satisfying demographic parity.
result Integer weights are optimal, allowing FairWASP to be understood as duplicating or eliminating samples.
The paper introduces return parity for fairness in MDPs, addressing delayed and adverse effects.
problem Fairness in MDPs for dynamic domains with delayed and adverse effects.
method Proposes return parity, decomposes return disparity, and develops algorithms for state visitation distributional alignment.
result The proposed algorithms can successfully close the disparity gap while maintaining policy performance.
In the context of machine learning, disparate impact refers to a form of systematic discrimination whereby the output distribution of a model depends on the value of a sensitive attribute (e.g., race or gender). In this paper, we propose an information-theoretic framework to analyze the disparate impact of a binary cla…
The increasing impact of algorithmic decisions on people's lives compels us to scrutinize their fairness and, in particular, the disparate impacts that ostensibly-color-blind algorithms can have on different groups. Examples include credit decisioning, hiring, advertising, criminal justice, personalized medicine, and t…
A transfer learning method builds high-dimensional models using disparate datasets.
problem Building comprehensive prediction models with small sample sizes and limited features.
method Transfer learning approach using external data to build a reduced model and apply calibration equations.
result Proposes a penalized generalized method of moment framework for inference and one-step estimation.
Study evaluates when splitting classifiers can improve performance despite disparate treatment.
problem Impact of disparate treatment in classification models.
method Comparison of split classifiers and group-blind classifiers, quantifying performance improvement.
result Proves an equivalent expression for the benefit-of-splitting which can be efficiently computed.
Adversarial training can lead to unfair accuracy disparities between different groups.
problem Adversarial training algorithms introduce unfair accuracy disparities between different groups of data.
method Propose a Fair-Robust-Learning (FRL) framework to mitigate unfairness in adversarial defenses.
result Empirical and theoretical validation of FRL's effectiveness in mitigating unfairness.
EL framework certifies and flags bias in ML models without distributional assumptions.
problem Systematic performance disparities across sensitive subpopulations in ML models.
method Empirical likelihood-based approach for non-parametric fairness auditing.
result EL framework outperforms bootstrap methods in certification and subpopulation discovery.
Study finds racial bias in pulse oximeter readings has minimal impact on ICU ventilation rates.
problem Racial disparities in pulse oximeter readings affect clinical decisions in ICU settings.
method Causal inference using path-specific effects and doubly robust estimator.
result Minimal impact of racial discrepancies on invasive ventilation rates, but more pronounced on ventilation duration.
Selective classification can worsen accuracy disparities between groups.
problem Selective classification can magnify existing accuracy disparities between various groups.
method Study of margin distribution and distributionally-robust models.
result Selective classification can uniformly improve each group on distributionally-robust models.
Selective regression allows abstention to improve fairness criteria.
problem Selective regression can exacerbate disparities between subgroups.
method Proposes new fairness criteria and two approaches to mitigate performance disparity.
result Proposed fairness criteria ensures performance improvement for every subgroup with reduced coverage.
Paper finds sharpness differences in transformer blocks accelerating LLM training.
problem Understanding and accelerating large language model pre-training.
method Uncovering sharpness disparity across transformer blocks and proposing Blockwise Learning Rate.
result Blockwise Learning Rate strategy accelerates LLM pre-training with lower loss and speedup.
Develops causal framework for fair survival analysis in healthcare.
problem Fairness in survival analysis for high-stakes domains like healthcare.
method Causal framework using graphical models, conditional survival function, and Causal Reduction Theorem.
result Decomposes disparities in survival into direct, indirect, and spurious pathways.
Develops methods for fair classification under linear disparity constraints.
problem Disparate impacts of machine learning algorithms on protected groups.
method Bayes-optimal fair classification methods via pre-, in-, and post-processing.
result Explicit forms of Bayes-optimal fair classifiers under linear disparity measures.
A membership inference attack (MIA) against a machine-learning model enables an attacker to determine whether a given data record was part of the model's training data or not. In this paper, we provide an in-depth study of the phenomenon of disparate vulnerability against MIAs: unequal success rate of MIAs against diff…
New framework reduces strategic manipulation cost for minority groups in fair classification.
problem Strategic manipulation disparities in fair classification.
method Constrained optimization framework that constructs classifiers to reduce strategic manipulation cost for minority groups.
result Empirically, the approach reduces strategic manipulation cost for minority groups over multiple real-world datasets.
In recent years, automated data-driven decision-making systems have enjoyed a tremendous success in a variety of fields (e.g., to make product recommendations, or to guide the production of entertainment). More recently, these algorithms are increasingly being used to assist socially sensitive decision-making (e.g., to…
Study decomposes racial healthcare disparities via shifts in mediator distributions.
problem Racial disparities in healthcare expenditures and their underlying drivers.
method Framework decomposing disparities into mediator distribution shifts and residual components, using MEPS data.
result Substantial disparities persist even when mediators are equalized, suggesting unmeasured or structural factors.
A new method for ILO with transition model disparity using an intermediary policy.
problem Learning tasks from expert observations with different transition dynamics.
method Training an intermediary policy to match the state transitions of the expert dataset.
result Our method outperforms existing ILO approaches with transition model mismatch.
Study breaks down graphs into structural and featural components for task-agnostic data valuation.
problem Lack of methods to assess the value of graphs in data marketplaces.
method Introduces blind message passing framework to evaluate graphs without specific task metrics.
result Demonstrates effectiveness in capturing structural disparities, relevance, and diversity of seller data for buyers.
Proposes a method to quantify and decompose disparity in ML models, separating exempt and non-exempt components.
problem Quantifying disparity in ML models, especially when certain features are exempted due to their critical importance.
method Information-theoretic decomposition into exempt and non-exempt components, satisfying desirable properties.
result Proposes a measure of non-exempt disparity that satisfies all desirable properties, and shows impossibility results for observational measures.
Standardized fairness measures for continuous risk scores using Wasserstein distance.
problem Quantifying and interpreting group disparities in continuous risk scores.
method Proposes standardized fairness measures based on Wasserstein distance for continuous scores.
result Proposed measures outperform ROC-based fairness measures by being more explicit and quantifying significant biases.
FOCUS addresses label quality disparity in FL for healthcare applications.
problem Label quality disparity in federated learning for healthcare applications.
method FOCUS maintains a small set of benchmark samples and computes the mutual cross-entropy between local and benchmark datasets to quantify label credibility. It then adjusts client weights based on credibility values.
result FOCUS effectively reduces the impact of noisy labels from clients, improving model performance.
Expands sparse disparity cues from LiDAR to improve stereo matching performance.
problem Improving stereo estimation performance with limited dense data.
method Proposes a sparsity expansion technique to enhance local features from sparse disparity cues.
result Significantly boosts stereo algorithms with sparse cues, outperforming previous methods.
Develops methods to measure and reduce fairness in datasets with limited protected attribute labels.
problem Measuring and reducing fairness in datasets with limited protected attribute labels.
method Proposes methods to estimate fairness metrics and train models to limit fairness violations using probabilistic protected attribute labels.
result Our methods provide tighter bounds on true disparity and effectively reduce fairness violations with lesser fairness-accuracy trade-offs.
We provide the asymptotic distribution of the major indexes used in the statistical literature to quantify disparate treatment in machine learning. We aim at promoting the use of confidence intervals when testing the so-called group disparate impact. We illustrate on some examples the importance of using confidence int…
The paper provides CI for test unfairness of group-fairness-aware classifiers trained with online SGD.
problem Ensuring fairness in machine learning models trained with stochastic gradient descent.
method Developed an online multiplier bootstrap method to estimate CI for test unfairness of DI and DM-aware linear classifiers.
result Asymptotic Central Limit Theorem holds for CI estimation of DI and DM-aware models.
Optimal pre-processing reduces disparate impact by minimizing total variation distance.
problem Achieving fairness in data outputs based on protected attributes.
method Using pre-processing to enforce fairness, minimizing total variation distance between pre-processed and original data distributions.
result The problem of fairness can be formulated as a linear program, efficiently solvable.
Assessing the fairness of a decision making system with respect to a protected class, such as gender or race, is challenging when class membership labels are unavailable. Probabilistic models for predicting the protected class based on observable proxies, such as surname and geolocation for race, are sometimes used to …
New method evaluates multiple social disparities using machine learning.
problem Reduction of educational disparities across multiple dimensions.
method Triply-Robust Machine Learning Approach for Causal Decomposition Analysis.
result Simultaneous interventions across multiple domains reduce disparities.
The paper proposes a fair reinforcement learning framework to prevent healthcare disparities.
problem Unfair reinforcement learning policies in healthcare can lead to socioeconomically-disadvantaged subgroups being underprivileged.
method The paper introduces a counterfactual fairness framework and a sequential data preprocessing algorithm to achieve fair sequential decision making.
result The proposed approach greatly enhances fair access to counseling in a digital health dataset designed to reduce opioid misuse.
Study highlights fairness issues in travel behavior prediction models.
problem Ethical challenges in travel behavior analysis using machine learning.
method Operationalized computational fairness by equality of opportunity; compared DNN and DCM; introduced absolute correlation regularization.
result Both DNN and DCM predict disparities across social groups, with DNN outperforming DCM in prediction disparities.
The paper shows how demographic data can lead to biased predictions, proposing 'Affirmative Information' as a solution.
problem Bias in predictions due to demographic data.
method Characterization of error types and conditions leading to disparate impact.
result Demographic variables in data can lead to biased predictions, with higher average outcomes receiving higher false positive rates.