Track papers across labs, topics, and trends.
Search papers, labs, and topics across Lattice.
To resolve persistent cold-start dynamics and item exhaustion in user-facing recommendation engines, this work introduces CohortMix-TS, a mixture bandit framework that warm-starts Thompson sampling by transferring latent user-group priors derived from prior cohorts' metadata. The system couples Bayesian prior transfer with constrained slate generation to mitigate rapid catalog depletion during local personalization. Evaluated in a 25-day live deployment with 713 participants, the approach significantly curtailed early cumulative regret and drove superior participant performance gains compared to standard exploration baselines.
Transferring latent group priors across cohorts eliminates the standard bandit cold-start penalty while actively preventing the premature catalog exhaustion that plagues real-world deployments.
Many recommender services repeatedly encounter cold-start cohorts, where new users arrive with little or no interaction history. This creates two challenges: learning user preferences quickly from limited feedback and sustaining useful recommendations when each user has a finite catalog that can become repetitive or depleted over time. We propose CohortMix-TS, a warm-started mixture bandit that learns latent user groups from earlier cohorts and uses available metadata to construct group-informed priors for new users. Starting from these fixed priors, the model personalizes independently as feedback from each user becomes available. Session slates combine Thompson sampling with diversity and inventory-depletion controls. We evaluate CohortMix-TS through simulation, semi-synthetic experiments, and a 25-day randomized in-the-wild deployment with 713 registered participants in a Campus Games quiz application. Our evaluations show that cross-cohort transfer improves early recommendation quality and user-level regret, while inventory-aware slate construction helps prevent premature exhaustion of preferred items. In the field deployment, treatment users also showed a larger early-to-late change in correctness than users receiving random recommendations. Together, these results show how warm-start transfer and inventory-aware recommendations can support personalization for short-lived, repeatedly cold-starting cohorts.