AI Paper Atlas

Research map / Bandits and online learning / Online convex optimization and games

Online Ad Allocation with Predictions

Fabian Christian Spaeh, Alina Ene

NeurIPS 2023 · Bandits and online learning › Online convex optimization and games · approach: Learning-augmented online algorithms · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
Use the learning-augmented algorithm instead of the worst-case algorithm without predictions, with 50% randomly corrupted OPT predictionsLearning-augmented Display Ads algorithm · iPinYourobustnessImproved
Use the algorithm with higher-quality predictors such as Previous Day or Dual BaseLearning-augmented Display Ads algorithm · iPinYou and Yahoorobustness vs worst-case algorithmImproved
Use the proposed algorithm instead of the random-mixture algorithm that interpolates between worst-case and predictionLearning-augmented Display Ads algorithm · all instancesperformanceImproved
Apply random and biased corruption to predictions on synthetic dataLearning-augmented Display Ads algorithm · SyntheticrobustnessNo clear effect
Use a Dual Base prediction built from only the first 10% of impressions (epsilon=0.1)Learning-augmented Display Ads algorithm · SyntheticperformanceImproved
Use one predictor (a single case) on real-world dataLearning-augmented Display Ads algorithm · real-world instancesperformance vs worst-caseHurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy