[2307.15142] Reconciling the accuracy-diversity trade-off in recommendations
Abstract:In recommendation settings, there is an apparent trade-off between the goals of accuracy (to recommend items a user is most likely to want) and diversity (to recommend items representing a range of categories). As such, real-world recommender systems often explicitly incorporate diversity separately from accuracy. This approach, however, leaves a basic question unanswered: Why is there a trade-off in the first place? We show how the trade-off can be explained via a user's consumption constraints -- users typically only consume a few of the items they are recommended. In a stylized model we introduce, objectives that account for this constraint induce diverse recommendations, while objectives that do not account for this constraint induce homogeneous recommendations. This suggests that accuracy and diversity appear misaligned because standard accuracy metrics do not consider consumption constraints. Our model yields precise and interpretable characterizations of diversity in different settings, giving practical insights into the design of diverse recommendations.
View PDF HTML (experimental) Abstract:In recommendation settings, there is an apparent trade-off between the goals of accuracy (to recommend items a user is most likely to want) and diversity (to recommend items representing a range of categories). As such, real-world recommender systems often explicitly incorporate diversity separately from accuracy. This approach, however, leaves a basic question unanswered: Why is there a trade-off in the first place? We show how the trade-off can be explained via a user's consumption constraints -- users typically only consume a few of the items they…
saved by
related reading
- Papers · Nikhil Garggargnikhil.com
- Shifting Consumption towards Diverse content via Reinforcement Learning | Spotify Researchresearch.atspotify.com
- Algorithmic Effects on the Diversity of Consumption on Spotify | Spotify Researchresearch.atspotify.com
- The_Use_MMR_Diversity_Based_LTMIR_1998.pdfcs.cmu.edu
- 2301.00993v1.pdfarxiv.org
- Bridging Systems: Open problems for countering destructive divisiveness across ranking, recommenders, and governance | Knight First Amendment Instituteknightcolumbia.org
- TikTok’s Secret Sauce | Knight First Amendment Instituteknightcolumbia.org
- [1904.06813] Personalized Re-ranking for Recommendationarxiv.org
- Effect of recommending users and opinions on the network connectivity and idea generation processarxiv.org
- Evaluating and Enhancing Robustness of Deep Recommendation Systems Against Hardware Errorsarxiv.org
- Netflix knows what you like - Digital Innovation and Transformationaiinstitute.hbs.edu
- the-algorithm/src/scala/com/twitter/simclusters_v2/README.md at main · twitter/the-algorithmgithub.com