aiqbench analysis
This paper treats fairness in recommenders as a bandit problem: it proves that policies ignoring user group membership accumulate linear regret over time, then proposes UCB-based variants that use group information. The trade-off is clear. Intersection-level UCB gives sublinear guarantees but scales exponentially with the number of groups, while the cheaper group-level heuristics work better only when a user's intersection preferences track their group preferences. Readers should note that the evidence rests on two datasets, Goodreads and BoardGameGeek, so generalization to other recommendation settings remains open.
Fair Recommender Systems for Intersecting Groups
arXiv:2610.40936v1 Announce Type: new Abstract: Recommendation systems that optimize for the average user can fail to capture the preferences of diverse demographic groups. We consider a dynamic recommendation system, where arriving users can belong to different demographic groups. Importantly, these groups can intersect, and a user may belong to several groups. We first show that over time, any group-blind policy (which excludes all current-user group information) recommending $K \geq 1$ items incurs linear total regret under mild conditions. In turn, we introduce three group-aware algorithms based on the classical Upper Confidence Bound algorithm (UCB): i) Intersection-Fitting, which runs UCB on each intersection, but has a sublinear group- and intersection-level regret guarantee with an exponential dependence in the number of groups; and ii) two group-level heuristics, Inner-Outer UCB and Group-Item UCB which allocate different arms to different groups, and use a sleeping expert meta-learner to choose which active group's recommendation an intersection should inherit at any given time step. We provide experimental results on two datasets, Goodreads and BoardGameGeek, and highlight that our heuristics can be preferable to the intersection-fitting algorithm when intersection-level preferences are closely related to group-level preferences, but that intersection-fitting performs better when different intersections within a same group prefer different items.
Abstract:Recommendation systems that optimize for the average user can fail to capture the preferences of diverse demographic groups. We consider a dynamic recommendation system, where arriving users can belong to different demographic groups. Importantly, these groups can intersect, and a user may belong to several groups. We first show that over time, any group-blind policy (which excludes all current-user group information) recommending $K \geq 1$ items incurs linear total regret under mild conditions. In turn, we introduce three group-aware algorithms based on the classical Upper Confidence Bound algorithm (UCB): i) Intersection-Fitting, which runs UCB on each intersection, but has a sublinear group- and intersection-level regret guarantee with an exponential dependence in the number of groups; and ii) two group-level heuristics, Inner-Outer UCB and Group-Item UCB which allocate different arms to different groups, and use a sleeping expert meta-learner to choose which active group's recommendation an intersection should inherit at any given time step. We provide experimental results on two datasets, Goodreads and BoardGameGeek, and highlight that our heuristics can be preferable to the intersection-fitting algorithm when intersection-level preferences are closely related to group-level preferences, but that intersection-fitting performs better when different intersections within a same group prefer different items.
Machine Learning (cs.LG); Computers and Society (cs.CY); Computer Science and Game Theory (cs.GT)
arXiv:2610.40936 [cs.LG]
(or arXiv:2610.40936v1 [cs.LG] for this version)
https://doi.org/10.48550/arXiv.2610.40936 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 303 words
Related coverage
Crash Narrative-Guided Countermeasure Recommendation Using Large Language Models: A Retrieval-Augmented Generation Framework for Intersection Safety
arXiv · Sep 16, 2026
GEM Training: How Meta Doubled the Efficiency of Its LLM-Scale Ads Foundation Model
Meta Engineering · Aug 3, 2026
Assessing the Role of Intersection Proximity in Pedestrian Crashes: Insights from Data Mining Approach
arXiv · Aug 7, 2026
The Null Is the Hard Part: Exact Tests for Memorization in Generative Models
arXiv · Oct 2, 2026