Calibrated Recommendations with Contextual Bandits

• URL: http://arxiv.org/abs/2509.05460v1
• Date: Fri, 05 Sep 2025 19:28:08 GMT
• Title: Calibrated Recommendations with Contextual Bandits
• Authors: Diego Feijer, Himan Abdollahpouri, Sanket Gupta, Alexander Clare, Yuxiao Wen, Todd Wasson, Maria Dimakopoulou, Zahra Nazari, Kyle Kretschman, Mounia Lalmas,
• Abstract summary: Spotify's Home page features a variety of content types, including music, podcasts, and audiobooks.<n>We propose a calibration method that leverages contextual bandits to learn each user's optimal content type distribution.<n>Our approach boosts engagement by adapting to how users interests in different content types varies across contexts.
• Score: 38.16600171259238
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Spotify's Home page features a variety of content types, including music, podcasts, and audiobooks. However, historical data is heavily skewed toward music, making it challenging to deliver a balanced and personalized content mix. Moreover, users' preference towards different content types may vary depending on the time of day, the day of week, or even the device they use. We propose a calibration method that leverages contextual bandits to dynamically learn each user's optimal content type distribution based on their context and preferences. Unlike traditional calibration methods that rely on historical averages, our approach boosts engagement by adapting to how users interests in different content types varies across contexts. Both offline and online results demonstrate improved precision and user engagement with the Spotify Home page, in particular with under-represented content types such as podcasts.

Related papers

• Content filtering methods for music recommendation: A review [1.104960878651584]
  This review focuses on the role of content filtering in mitigating biases inherent in collaborative filtering approaches.<n>We explore various methods of song classification for content filtering, including lyrical analysis using Large Language Models (LLMs) and audio signal processing techniques.
  arXiv  Detail & Related papers  (2025-07-03T03:44:20Z)
• Enhancing Sequential Music Recommendation with Personalized Popularity Awareness [56.972624411205224]
  This paper introduces a novel approach that incorporates personalized popularity information into sequential recommendation. Experimental results demonstrate that a Personalized Most Popular recommender outperforms existing state-of-the-art models.
  arXiv  Detail & Related papers  (2024-09-06T15:05:12Z)
• Ranking Across Different Content Types: The Robust Beauty of Multinomial Blending [1.0650780147044159]
  Cross-content-type ranking poses a significant challenge for traditional learning-to-rank algorithms. We explore a simple method for cross-content-type ranking, called multinomial blending (MB) We report the results of an A/B test from an Amazon Music ranking use-case.
  arXiv  Detail & Related papers  (2024-08-17T11:11:31Z)
• Fairness Through Domain Awareness: Mitigating Popularity Bias For Music Discovery [56.77435520571752]
  We explore the intrinsic relationship between music discovery and popularity bias. We propose a domain-aware, individual fairness-based approach which addresses popularity bias in graph neural network (GNNs) based recommender systems. Our approach uses individual fairness to reflect a ground truth listening experience, i.e., if two songs sound similar, this similarity should be reflected in their representations.
  arXiv  Detail & Related papers  (2023-08-28T14:12:25Z)
• Improving Recommendation System Serendipity Through Lexicase Selection [53.57498970940369]
  We propose a new serendipity metric to measure the presence of echo chambers and homophily in recommendation systems. We then attempt to improve the diversity-preservation qualities of well known recommendation techniques by adopting a parent selection algorithm known as lexicase selection. Our results show that lexicase selection, or a mixture of lexicase selection and ranking, outperforms its purely ranked counterparts in terms of personalization, coverage and our specifically designed serendipity benchmark.
  arXiv  Detail & Related papers  (2023-05-18T15:37:38Z)
• Like a Good Nearest Neighbor: Practical Content Moderation and Text Classification [66.02091763340094]
  Like a Good Nearest Neighbor (LaGoNN) is a modification to SetFit that introduces no learnable parameters but alters input text with information from its nearest neighbor. LaGoNN is effective at flagging undesirable content and text classification, and improves the performance of SetFit.
  arXiv  Detail & Related papers  (2023-02-17T15:43:29Z)
• A Dataset for Greek Traditional and Folk Music: Lyra [69.07390994897443]
  This paper presents a dataset for Greek Traditional and Folk music that includes 1570 pieces, summing in around 80 hours of data. The dataset incorporates YouTube timestamped links for retrieving audio and video, along with rich metadata information with regards to instrumentation, geography and genre.
  arXiv  Detail & Related papers  (2022-11-21T14:15:43Z)
• Preference Dynamics Under Personalized Recommendations [12.89628003097857]
  We show whether some phenomenon akin to polarization occurs when users receive personalized content recommendations. A more interesting objective is to understand under what conditions a recommendation algorithm can ensure stationarity of user's preferences.
  arXiv  Detail & Related papers  (2022-05-25T19:29:53Z)
• Follow the guides: disentangling human and algorithmic curation in online music consumption [1.4506962780822348]
  We analyze the complete listening history of about 9k users over one year. We show that the two types of recommendation offered by music platforms -- algorithmic and editorial -- may drive the consumption of more or less diverse content in opposite directions.
  arXiv  Detail & Related papers  (2021-09-08T20:14:48Z)
• Recommending Podcasts for Cold-Start Users Based on Music Listening and Taste [5.429958676933934]
  We consider podcasting as an emerging medium with rapid growth in adoption. Using music consumption behavior, we examine two main techniques in inferring Spotify users preferences over more than 200k podcasts. Our results show significant improvements in consumption of up to 50% for both offline and online experiments.
  arXiv  Detail & Related papers  (2020-07-27T02:55:23Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.