Fairness-Aware Ranking in Search & Recommendation Systems with Application to LinkedIn Talent Search

Abstract

We present a framework for quantifying and mitigating algorithmic bias in mechanisms designed for ranking individuals, typically used as part of web-scale search and recommendation systems. We first propose complementary measures to quantify bias with respect to protected attributes such as gender and age. We then present algorithms for computing fairness-aware re-ranking of results. For a given search or recommendation task, our algorithms seek to achieve a desired distribution of top ranked results with respect to one or more protected attributes. We show that such a framework can be tailored to achieve fairness criteria such as equality of opportunity and demographic parity depending on the choice of the desired distribution. We evaluate the proposed algorithms via extensive simulations over different parameter choices, and study the effect of fairness-aware ranking on both bias and utility measures. We finally present the online A/B testing results from applying our framework towards representative ranking in LinkedIn Talent Search, and discuss the lessons learned in practice. Our approach resulted in tremendous improvement in the fairness metrics (nearly three fold increase in the number of search queries with representative results) without affecting the business metrics, which paved the way for deployment to 100% of LinkedIn Recruiter users worldwide. Ours is the first large-scale deployed framework for ensuring fairness in the hiring domain, with the potential positive impact for more than 630M LinkedIn members.

arXiv commentThis paper has been accepted for publication at ACM KDD 2019

References in corpus (10)

• Semantics derived automatically from language corpora contain human-like biases
• Equality of Opportunity in Supervised Learning
• Man is to Computer Programmer as Woman is to Homemaker? Debiasing Word Embeddings
• Fairness Beyond Disparate Treatment & Disparate Impact: Learning Classification without Disparate Mistreatment
• Fairness of Exposure in Rankings
• FA*IR: A Fair Top-k Ranking Algorithm
• Equity of Attention: Amortizing Individual Fairness in Rankings
• On the (im)possibility of fairness
• Designing Fair Ranking Schemes
• How to be Fair and Diverse?

Cited by in corpus (77)

• Mitigating Bias in Algorithmic Hiring: Evaluating Claims and Practices
• A Survey on the Fairness of Recommender Systems
• Fairness in Recommender Systems: Research Landscape and Future Directions
• Towards Long-term Fairness in Recommendation
• FairRec: Two-Sided Fairness for Personalized Recommendations in Two-Sided Platforms
• Building Human Values into Recommender Systems: An Interdisciplinary Synthesis
• Fairness and Bias in Algorithmic Hiring: a Multidisciplinary Survey
• Toward Pareto Efficient Fairness-Utility Trade-off inRecommendation through Reinforcement Learning
• A Graph-based Approach for Mitigating Multi-sided Exposure Bias in Recommender Systems
• User Fairness, Item Fairness, and Diversity for Rankings in Two-Sided Markets
• Explainable Fairness in Recommendation
• Societal Biases in Retrieved Contents: Measurement Framework and Adversarial Mitigation for BERT Rankers
• FairSR: Fairness-aware Sequential Recommendation through Multi-Task Learning with Preference Graph Embeddings
• LiFT: A Scalable Framework for Measuring Fairness in ML Applications
• Evaluation Metrics for Measuring Bias in Search Engine Results
• A Model-Agnostic Causal Learning Framework for Recommendation using Search Data
• Amazon SageMaker Clarify: Machine Learning Bias Detection and Explainability in the Cloud
• Towards Fair Recommendation in Two-Sided Platforms
• Socially Responsible AI Algorithms: Issues, Purposes, and Challenges
• Deep Learning on Knowledge Graph for Recommender System: A Survey
• Policy-Gradient Training of Fair and Unbiased Ranking Functions
• Exploring ChatGPT for Next-generation Information Retrieval: Opportunities and Challenges
• Towards Individual and Multistakeholder Fairness in Tourism Recommender Systems
• De-biasing "bias" measurement
• Fairness of Exposure in Light of Incomplete Exposure Estimation
• Understanding and Mitigating the Effect of Outliers in Fair Ranking
• Intersectional Two-sided Fairness in Recommendation
• Lagrangian Inference for Ranking Problems
• Fairness-Aware Explainable Recommendation over Knowledge Graphs
• FairCanary: Rapid Continuous Explainable Fairness
• Evaluating Fairness in Argument Retrieval
• FAIR: Fairness-Aware Information Retrieval Evaluation
• Learning for Counterfactual Fairness from Observational Data
• Cognitively Biased Users Interacting with Algorithmically Biased Results in Whole-Session Search on Debated Topics
• Introducing the Expohedron for Efficient Pareto-optimal Fairness-Utility Amortizations in Repeated Rankings
• Subverting Fair Image Search with Generative Adversarial Perturbations
• Towards Fairness in Personalized Ads Using Impression Variance Aware Reinforcement Learning
• Path-Specific Counterfactual Fairness for Recommender Systems
• The Role of Relevance in Fair Ranking
• Unmasking Gender Bias in Recommendation Systems and Enhancing Category-Aware Fairness
• Characterizing Intersectional Group Fairness with Worst-Case Comparisons
• Social Norm Bias: Residual Harms of Fairness-Aware Algorithms
• Probabilistic Permutation Graph Search: Black-Box Optimization for Fairness in Ranking
• Predictive Uncertainty-based Bias Mitigation in Ranking
• Evaluating the Fairness of Discriminative Foundation Models in Computer Vision
• Representation Online Matters: Practical End-to-End Diversification in Search and Recommender Systems
• Properties of Group Fairness Metrics for Rankings
• On the Problem of Underranking in Group-Fair Ranking
• A Framework for Fairness in Two-Sided Marketplaces
• FARA: Future-aware Ranking Algorithm for Fairness Optimization
• When the Umpire is also a Player: Bias in Private Label Product Recommendations on E-commerce Marketplaces
• Automatic Feature Fairness in Recommendation via Adversaries
• Recommendation Fairness: From Static to Dynamic
• Intersectional Affirmative Action Policies for Top-k Candidates Selection
• Maximizing Neutrality in News Ordering
• Fairness Through Regularization for Learning to Rank
• Mapping Stakeholder Needs to Multi-Sided Fairness in Candidate Recommendation for Algorithmic Hiring
• Social Perception of Faces in a Vision-Language Model
• Individually Fair Ranking
• hyperFA*IR: A hypergeometric approach to fair rankings with finite candidate pool
• Countering Mainstream Bias via End-to-End Adaptive Local Learning
• Measuring Fairness Under Unawareness of Sensitive Attributes: A Quantification-Based Approach
• Fairness-Aware Online Personalization
• Calibrating Explore-Exploit Trade-off for Fair Online Learning to Rank
• Maxmin-Fair Ranking: Individual Fairness under Group-Fairness Constraints
• Towards More Accountable Search Engines: Online Evaluation of Representation Bias
• FAST: A Fairness Assured Service Recommendation Strategy Considering Service Capacity Constraint
• Towards Fair Rankings: Leveraging LLMs for Gender Bias Detection and Measurement
• Problem Learning: Towards the Free Will of Machines
• Robust Allocations with Diversity Constraints
• A Sociotechnical View of Algorithmic Fairness
• Incentives for Item Duplication under Fair Ranking Policies
• HyperFair: A Soft Approach to Integrating Fairness Criteria
• Generative Ghost: Investigating Ranking Bias Hidden in AI-Generated Videos
• Facets of Fairness in Search and Recommendation
• Practical Compositional Fairness: Understanding Fairness in Multi-Component Recommender Systems
• Job Recommender Systems: A Review