The Disparate Benefits of Deep Ensembles

• URL: http://arxiv.org/abs/2410.13831v1
• Date: Thu, 17 Oct 2024 17:53:01 GMT
• Title: The Disparate Benefits of Deep Ensembles
• Authors: Kajetan Schweighofer, Adrian Arnaiz-Rodriguez, Sepp Hochreiter, Nuria Oliver,
• Abstract summary: We investigate the interplay between performance gains from Deep Ensembles and fairness. We find that they unevenly favor different groups in what we refer to as a disparate benefits effect. Our findings show that post-processing is an effective method to mitigate this unfairness.
• Score: 11.303233667605586
• License:
• Abstract: Ensembles of Deep Neural Networks, Deep Ensembles, are widely used as a simple way to boost predictive performance. However, their impact on algorithmic fairness is not well understood yet. Algorithmic fairness investigates how a model's performance varies across different groups, typically defined by protected attributes such as age, gender, or race. In this work, we investigate the interplay between the performance gains from Deep Ensembles and fairness. Our analysis reveals that they unevenly favor different groups in what we refer to as a disparate benefits effect. We empirically investigate this effect with Deep Ensembles applied to popular facial analysis and medical imaging datasets, where protected group attributes are given and find that it occurs for multiple established group fairness metrics, including statistical parity and equal opportunity. Furthermore, we identify the per-group difference in predictive diversity of ensemble members as the potential cause of the disparate benefits effect. Finally, we evaluate different approaches to reduce unfairness due to the disparate benefits effect. Our findings show that post-processing is an effective method to mitigate this unfairness while preserving the improved performance of Deep Ensembles.

Related papers

• Positive-Sum Fairness: Leveraging Demographic Attributes to Achieve Fair AI Outcomes Without Sacrificing Group Gains [0.0]
  We argue that decreases in fairness can be harmful or non-harmful depending on the type of change and how sensitive attributes are used. We introduce the notion of positive-sum fairness, which states that an increase in performance that results in a larger group disparity is acceptable as long as it does not come at the cost of individual subgroup performance.
  arXiv  Detail & Related papers  (2024-09-30T04:37:23Z)
• GroupMixNorm Layer for Learning Fair Models [4.324785083027206]
  This research proposes a novel in-processing based GroupMixNorm layer for mitigating bias from deep learning models. The proposed method improves upon several fairness metrics with minimal impact on overall accuracy.
  arXiv  Detail & Related papers  (2023-12-19T09:04:26Z)
• Counterpart Fairness -- Addressing Systematic between-group Differences in Fairness Evaluation [17.495053606192375]
  When using machine learning to aid decision-making, it is critical to ensure that an algorithmic decision is fair and does not discriminate against specific individuals/groups. Existing group fairness methods aim to ensure equal outcomes across groups delineated by protected variables like race or gender. In cases where systematic differences between groups play a significant role in outcomes, these methods may overlook the influence of non-protected variables.
  arXiv  Detail & Related papers  (2023-05-29T15:41:12Z)
• Fair Effect Attribution in Parallel Online Experiments [57.13281584606437]
  A/B tests serve the purpose of reliably identifying the effect of changes introduced in online services. It is common for online platforms to run a large number of simultaneous experiments by splitting incoming user traffic randomly. Despite a perfect randomization between different groups, simultaneous experiments can interact with each other and create a negative impact on average population outcomes.
  arXiv  Detail & Related papers  (2022-10-15T17:15:51Z)
• Fairness via Adversarial Attribute Neighbourhood Robust Learning [49.93775302674591]
  We propose a principled underlineRobust underlineAdversarial underlineAttribute underlineNeighbourhood (RAAN) loss to debias the classification head.
  arXiv  Detail & Related papers  (2022-10-12T23:39:28Z)
• Fair Bayes-Optimal Classifiers Under Predictive Parity [33.648053823193855]
  This paper considers predictive parity, which requires equalizing the probability of success given a positive prediction among different protected groups. We propose an algorithm we call FairBayes-DPP, aiming to ensure predictive parity when our condition is satisfied.
  arXiv  Detail & Related papers  (2022-05-15T04:58:10Z)
• Cascaded Debiasing: Studying the Cumulative Effect of Multiple Fairness-Enhancing Interventions [48.98659895355356]
  This paper investigates the cumulative effect of multiple fairness enhancing interventions at different stages of the machine learning (ML) pipeline. Applying multiple interventions results in better fairness and lower utility than individual interventions on aggregate. On the downside, fairness-enhancing interventions can negatively impact different population groups, especially the privileged group.
  arXiv  Detail & Related papers  (2022-02-08T09:20:58Z)
• MultiFair: Multi-Group Fairness in Machine Learning [52.24956510371455]
  We study multi-group fairness in machine learning (MultiFair) We propose a generic end-to-end algorithmic framework to solve it. Our proposed framework is generalizable to many different settings.
  arXiv  Detail & Related papers  (2021-05-24T02:30:22Z)
• Deconfounding Scores: Feature Representations for Causal Effect Estimation with Weak Overlap [140.98628848491146]
  We introduce deconfounding scores, which induce better overlap without biasing the target of estimation. We show that deconfounding scores satisfy a zero-covariance condition that is identifiable in observed data. In particular, we show that this technique could be an attractive alternative to standard regularizations.
  arXiv  Detail & Related papers  (2021-04-12T18:50:11Z)
• Interpretable Assessment of Fairness During Model Evaluation [1.2183405753834562]
  We introduce a novel hierarchical clustering algorithm to detect heterogeneity among users in given sets of sub-populations. We demonstrate the performance of the algorithm on real data from LinkedIn.
  arXiv  Detail & Related papers  (2020-10-26T02:31:17Z)
• Almost-Matching-Exactly for Treatment Effect Estimation under Network Interference [73.23326654892963]
  We propose a matching method that recovers direct treatment effects from randomized experiments where units are connected in an observed network. Our method matches units almost exactly on counts of unique subgraphs within their neighborhood graphs.
  arXiv  Detail & Related papers  (2020-03-02T15:21:20Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.