Filters








393,549 Hits in 5.9 sec

Maxmin-Fair Ranking: Individual Fairness under Group-Fairness Constraints [article]

David Garcia-Soriano, Francesco Bonchi
2021 arXiv   pre-print
We study a novel problem of fairness in ranking aimed at minimizing the amount of individual unfairness introduced when enforcing group-fairness constraints.  ...  , while satisfying the given group-fairness constraints, ensure that the maximum possible value is brought to individuals.  ...  MAXMIN-FAIR RANKING We are given a set of individuals to be ranked U, a partition of U into groups 1 , . . . , , and a relevance function : U → R.  ... 
arXiv:2106.08652v2 fatcat:g6wytj6pv5cq7c6sqf7tfax5pm

On the Problem of Underranking in Group-Fair Ranking [article]

Sruthi Gorantla, Amit Deshpande, Anand Louis
2021 arXiv   pre-print
Most group-fair ranking algorithms post-process a given ranking and output a group-fair ranking.  ...  We give a fair ranking algorithm that takes any given ranking and outputs another ranking with simultaneous underranking and group fairness guarantees comparable to the lower bound we prove.  ...  That is, the output group fair ranking has strictly less than 1 individual fairness. This manifests the trade-off between the group fairness and the individual fairness in ranking.  ... 
arXiv:2010.06986v2 fatcat:usbxdgk6gnfvjiszyywwqb55be

Fair Ranking as Fair Division: Impact-Based Individual Fairness in Ranking [article]

Yuta Saito, Thorsten Joachims
2022 pre-print
Beyond these theoretical results, we illustrate empirically how our framework controls the trade-off between impact-based individual item fairness and user utility.  ...  To compute ranking policies that are fair according to these axioms, we propose a new ranking objective related to the Nash Social Welfare.  ...  individual item fairness.  ... 
doi:10.1145/3534678.3539353 arXiv:2206.07247v1 fatcat:cbqxer3glfhtbkr3322qwvskxi

Equity of Attention

Asia J. Biega, Krishna P. Gummadi, Gerhard Weikum
2018 The 41st International ACM SIGIR Conference on Research & Development in Information Retrieval - SIGIR '18  
Second, as no single ranking can achieve individual attention fairness, we propose a novel mechanism that achieves amortized fairness, where attention accumulated across a series of rankings is proportional  ...  Our experimental evaluation reveals that unfair attention distribution in rankings can be substantial, and demonstrates that our method can improve individual fairness while retaining high ranking quality  ...  Observe that this modified fairness definition allows us to permute individual rankings so as to satisfy fairness requirements over time.  ... 
doi:10.1145/3209978.3210063 dblp:conf/sigir/BiegaGW18 fatcat:at2cbbwlmbdhlfk3kndzekhecm

FairSight: Visual Analytics for Fairness in Decision Making

Yongsu Ahn, Yu-Ru Lin
2019 IEEE Transactions on Visualization and Computer Graphics  
We propose FairSight, a visual analytic system to address this need; it is designed to achieve different notions of fairness in ranking decisions through identifying the required actions - understanding  ...  more fair outcomes.  ...  In Ranking List View, she was able to compare all generated rankings with Group fairness, Individual fairness, and Utiliy measures.  ... 
doi:10.1109/tvcg.2019.2934262 pmid:31425083 fatcat:njxavllstjdwzfqkbsxkr5marq

iFair: Learning Individually Fair Data Representations for Algorithmic Decision Making [article]

Preethi Lahoti, Krishna P. Gummadi, Gerhard Weikum
2019 arXiv   pre-print
The paper introduces a method for probabilistically mapping user records into a low-rank representation that reconciles individual fairness and the utility of classifiers and rankings in downstream applications  ...  In contrast, the alternative paradigm of individual fairness has received relatively little attention, and this paper advances this less explored direction.  ...  This research was supported by the ERC Synergy Grant "imPACT" (No. 610150) and ERC Advanced Grant "Foundations for Fair Social Computing" (No. 789373).  ... 
arXiv:1806.01059v2 fatcat:iqci7fthbngjlg5ccpiydcwxny

How Fair is Fairness-aware Representative Ranking and Methods for Fair Ranking [article]

Akrati Saxena, George Fletcher, Mykola Pechenizkiy
2021 arXiv   pre-print
We define individual unfairness and group unfairness and propose methods to generate ideal individual and group fair representative ranking if the universal representation ratio is known or unknown.  ...  In this work, we highlight the bias in fairness-aware representative ranking for an individual as well as for a group if the group is sub-active on the platform.  ...  Individual Fair Representative Ranking A ranking is fair for an individual if ∈ , , then ∈ , , .  ... 
arXiv:2103.01335v1 fatcat:zdmn2sxetzgabaxnxliciyynxi

Measuring Fairness in Ranked Outputs [article]

Ke Yang, Julia Stoyanovich
2016 arXiv   pre-print
This warrants a careful study of the fairness of a ranking scheme. In this paper we propose fairness measures for ranked outputs.  ...  The final output is a ranking that represents the relative quality of the individuals.  ...  A useful dichotomy is between individual fairness -a requirement that similar individuals are treated similarly, and group fairness, also known as statistical parity -a requirement that demographics of  ... 
arXiv:1610.08559v1 fatcat:hln73vjxqfcmtl5uypaflb45gi

Effective Exposure Amortizing for Fair Top-k Recommendation [article]

Tao Yang, Zhichao Xu, Qingyao Ai
2022 arXiv   pre-print
individual level and group level.  ...  Thus, how to maintain a balance between ranking relevance and fairness is important to both producers and customers.  ...  Individual fairness datasets are used for evaluating methods for individual fairness and they don't contain any groups.  ... 
arXiv:2204.03046v1 fatcat:c7nr2vnm7resdarxucfy6lzkrq

Facets of Fairness in Search and Recommendation [article]

Sahil Verma, Ruoyuan Gao, Chirag Shah
2020 arXiv   pre-print
Then, it focuses on explaining the emerging concept of fairness in various recommendation settings.  ...  Counteracting this bias and bringing a certain amount of fairness in search is crucial to not only creating a more balanced environment that considers relevance and diversity but also providing a more  ...  The ranker does not account for individual preferences of the consumers of the ranked list. The fairness in this setting addresses how the candidates are ranked.  ... 
arXiv:2008.01194v1 fatcat:vtmdj65nvndbbavaot6nyii2qi

Policy Learning for Fairness in Ranking [article]

Ashudeep Singh, Thorsten Joachims
2019 arXiv   pre-print
Furthermore, we provide a new LTR algorithm called Fair-PG-Rank for directly searching the space of fair ranking policies via a policy-gradient approach.  ...  This framework expands the class of learnable ranking functions to stochastic ranking policies, which provides a language for rigorously expressing fairness specifications.  ...  ranking functions under both individual fairness and group fairness constraints.  ... 
arXiv:1902.04056v2 fatcat:gc7tdm5mtjhgfhtu6hc6hg7l6y

A Survey on the Fairness of Recommender Systems

Yifan Wang, Weizhi Ma, Min Zhang, Yiqun Liu, Shaoping Ma
2022 ACM Transactions on Information Systems  
First, we summarize fairness definitions in the recommendation and provide several views to classify fairness issues.  ...  However, due to multiple complex resource allocation processes and various fairness definitions, the research on fairness in recommendation is scattered.  ...  Individual fairness believes that outcomes should be fair at the individual level. Individual fairness in some work refers to the idea that similar individuals should be treated similarly [7, 21] .  ... 
doi:10.1145/3547333 fatcat:uztccsrtxzcwfna56qxnuuj6pe

A Survey on the Fairness of Recommender Systems [article]

Yifan Wang, Weizhi Ma, Min Zhang, Yiqun Liu, Shaoping Ma
2022 arXiv   pre-print
First, we summarize fairness definitions in the recommendation and provide several views to classify fairness issues.  ...  However, due to multiple complex resource allocation processes and various fairness definitions, the research on fairness in recommendation is scattered.  ...  Individual fairness believes that outcomes should be fair at the individual level. Individual fairness in some work refers to the idea that similar individuals should be treated similarly [7, 23] .  ... 
arXiv:2206.03761v2 fatcat:o2xnm4o6qrfglbmlt3mdxq5d6u

A Nutritional Label for Rankings

Ke Yang, Julia Stoyanovich, Abolfazl Asudeh, Bill Howe, HV Jagadish, Gerome Miklau
2018 Proceedings of the 2018 International Conference on Management of Data - SIGMOD '18  
Ranking Facts is made up of a collection of visual widgets that implement our latest research results on fairness, stability, and transparency for rankings, and that communicate details of the ranking  ...  Similar concerns apply in cases where items other than individuals are ranked, including colleges, academic departments, or products.  ...  individuals [13] .  ... 
doi:10.1145/3183713.3193568 dblp:conf/sigmod/YangSAHJM18 fatcat:csradeztznbxjc25rn5jt55axe

Joint Multisided Exposure Fairness for Recommendation [article]

Haolun Wu, Bhaskar Mitra, Chen Ma, Fernando Diaz, Xue Liu
2022 arXiv   pre-print
Prior research on exposure fairness in the context of recommender systems has focused mostly on disparities in the exposure of individual or groups of items to individual users of the system.  ...  Furthermore, we study and discuss the relationships between the different exposure fairness dimensions proposed in this paper, as well as demonstrate how stochastic ranking policies can be optimized towards  ...  We refer to this as individual-user to individual-item fairness, or II-F to be more concise.  ... 
arXiv:2205.00048v1 fatcat:mrvxm2ryfvfhvnskqgkq7qcx5a
« Previous Showing results 1 — 15 out of 393,549 results