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

Interference is a ubiquitous problem in experiments conducted on two-sidedcontent marketplaces, such as Douyin (China's analog of TikTok). In many cases,creators are the natural unit of experimentation, but creators interfere witheach other through competition for viewers' limited time and attention. "Naive"estimators currently used in practice simply ignore the interference, but indoing so incur bias on the order of the treatment effect. We formalize theproblem of inference in such experiments as one of policy evaluation.Off-policy estimators, while unbiased, are impractically high variance. Weintroduce a novel Monte-Carlo estimator, based on "Differences-in-Qs" (DQ)techniques, which achieves bias that is second-order in the treatment effect,while remaining sample-efficient to estimate. On the theoretical side, ourcontribution is to develop a generalized theory of Taylor expansions for policyevaluation, which extends DQ theory to all major MDP formulations. On thepractical side, we implement our estimator on Douyin's experimentationplatform, and in the process develop DQ into a truly "plug-and-play" estimatorfor interference in real-world settings: one which provides robust, low-bias,low-variance treatment effect estimates; admits computationally cheap,asymptotically exact uncertainty quantification; and reduces MSE by 99\%compared to the best existing alternatives in our applications.

Correcting for Interference in Experiments: A Case Study at Douyin