A dashboard for controlling polarization in personalization

Celis, L. Elisa; Kapoor, Sayash; Salehi, Farnood; Keswani, Vijay; Vishnoi, Nisheeth K.

doi:10.3233/aic-180606

Cited by 4 publications

(3 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Tools have also been developed to help users control the results of recommendation algorithms, to surface and adjust for bias that might emerge. One such tool was presented at the IJCAI-ECAI-18 Demonstration Track for controlling polarisation in news feeds [60].…”

Section: Accountability In Aimentioning

confidence: 99%

Accountability in AI: From Principles to Industry-specific Accreditation

Percy¹,

Dragičević²,

Sarkar³

et al. 2021

Preprint

View full text Add to dashboard Cite

Recent AI-related scandals have shed a spotlight on accountability in AI, with increasing public interest and concern. This paper draws on literature from public policy and governance to make two contributions. First, we propose an AI accountability ecosystem as a useful lens on the system, with different stakeholders requiring and contributing to specific accountability mechanisms. We argue that the present ecosystem is unbalanced, with a need for improved transparency via AI explainability and adequate documentation and process formalisation to support internal audit, leading up eventually to external accreditation processes. Second, we use a case study in the gambling sector to illustrate in a subset of the overall ecosystem the need for industry-specific accountability principles and processes. We define and evaluate critically the implementation of key accountability principles in the gambling industry, namely addressing algorithmic bias and model explainability, before concluding and discussing directions for future work based on our findings.

show abstract

Section: Accountability In Aimentioning

confidence: 99%

Accountability in AI: From Principles to Industry-specific Accreditation

Percy¹,

Dragičević²,

Sarkar³

et al. 2021

Preprint

View full text Add to dashboard Cite

show abstract

“…Several prior works address the problems of polarization and algorithmic bias, including [10,24] who control polarization in social-networks and personalized feeds, [42] who diversify personal feeds, and [12] who create a diverse and balanced summary of a set of results. In addition, [6,13,44] study fair ranking algorithms; these could be used to generate a balanced list of results on job platforms and other search engines.…”

Section: Optimization Problemmentioning

confidence: 99%

“…Advertiser i's bids are being increased faster than or equal to any other advertiser's. Recalling that the shift of advertiser n, α ij = 0 for all user types j ∈ [m], using Equation (10) we can express q ij (tu + α) and its gradient as shown in Figure 6.…”

Section: Proof Of Lemma 73mentioning

confidence: 99%

Toward Controlling Discrimination in Online Ad Auctions

Celis,

Mehrotra,

Vishnoi

2019

Preprint

Self Cite

View full text Add to dashboard Cite

Online advertising platforms are thriving due to the customizable audiences they offer advertisers. However, recent studies show that advertisements can be discriminatory with respect to the gender or race of the audience that sees the ad, and may inadvertently cross ethical and/or legal boundaries. To prevent this, we propose a constrained ad auction framework that maximizes the platforms revenue conditioned on ensuring that the audience seeing an advertisers ad is distributed appropriately across sensitive types such as gender or race. Building upon Myersons classic work, we first present an optimal auction mechanism for a large class of fairness constraints. Finding the parameters of this optimal auction, however, turns out to be a non-convex problem. We show that this non-convex problem can be reformulated as a more structured non-convex problem with no saddle points or local-maxima; this allows us to develop a gradient-descent-based algorithm to solve it. Our empirical results on the A1 Yahoo! dataset demonstrate that our algorithm can obtain uniform coverage across different user types for each advertiser at a minor loss to the revenue of the platform, and a small change to the size of the audience each advertiser reaches.

show abstract

Eliminating Latent Discrimination: Train Then Mask

Ghili

Kazemi²,

Karbasi

2019

SSRN Journal

View full text Add to dashboard Cite

How can we control for latent discrimination in predictive models? How can we provably remove it? Such questions are at the heart of algorithmic fairness and its impacts on society. In this paper, we define a new operational fairness criteria, inspired by the well-understood notion of omitted variable-bias in statistics and econometrics. Our notion of fairness effectively controls for sensitive features and provides diagnostics for deviations from fair decision making. We then establish analytical and algorithmic results about the existence of a fair classifier in the context of supervised learning. Our results readily imply a simple, but rather counter-intuitive, strategy for eliminating latent discrimination. In order to prevent other features proxying for sensitive features, we need to include sensitive features in the training phase, but exclude them in the test/evaluation phase while controlling for their effects. We evaluate the performance of our algorithm on several real-world datasets and show how fairness for these datasets can be improved with a very small loss in accuracy.

show abstract

A dashboard for controlling polarization in personalization

Cited by 4 publications

References 15 publications

Accountability in AI: From Principles to Industry-specific Accreditation

Accountability in AI: From Principles to Industry-specific Accreditation

Toward Controlling Discrimination in Online Ad Auctions

Eliminating Latent Discrimination: Train Then Mask

Contact Info

Product

Resources

About