English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Recommender systems operate in an inherently dynamical setting. Pastrecommendations influence future behavior, including which data points areobserved and how user preferences change. However, experimenting in productionsystems with real user dynamics is often infeasible, and existingsimulation-based approaches have limited scale. As a result, manystate-of-the-art algorithms are designed to solve supervised learning problems,and progress is judged only by offline metrics. In this work we investigate theextent to which offline metrics predict online performance by evaluating elevenrecommenders across six controlled simulated environments. We observe thatoffline metrics are correlated with online performance over a range ofenvironments. However, improvements in offline metrics lead to diminishingreturns in online performance. Furthermore, we observe that the ranking ofrecommenders varies depending on the amount of initial offline data available.We study the impact of adding exploration strategies, and observe that theireffectiveness, when compared to greedy recommendation, is highly dependent onthe recommendation algorithm. We provide the environments and recommendersdescribed in this paper as Reclab: an extensible ready-to-use simulationframework at https://github.com/berkeley-reclab/RecLab.

Do Offline Metrics Predict Online Performance in Recommender Systems?