Research map / Bandits and online learning / Online convex optimization and games
Improved Online Learning Algorithms for CTR Prediction in Ad Auctions
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 changed | Where | Measured | Effect |
|---|---|---|---|
| Use a UCB-style stage-IC auction for myopic advertisers | Online pay-per-click ad auctions · Adversarial valuations | regret: \widetilde{O}(\sqrt{T}) | Helped |
| Use UCB estimates for CTRs in the stage-IC auction | Online pay-per-click ad auctions · Static valuations with a time-independent positive gap | regret: -0.05 \zeta T + O(\log(nT)) | Helped |
| Add an explore-then-commit phase with free round-robin exploration, then commit using UCB-based auction scores | Online pay-per-click ad auctions with non-myopic advertisers · Static valuations with a time-independent positive gap | regret: -\Omega(T) | Helped |
| Use a prior global-IC explore-then-commit mechanism | Online pay-per-click ad auctions with non-myopic advertisers · Adversarial valuations | regret: O(T^{2/3}) | Hurt |
| Require only stage-IC and individual rationality for the mechanism class | Online ad auctions · Fixed valuations with gap of order 1/\sqrt{T} | regret lower bound: \Omega(\sqrt{T}) | Hurt |
Closest papers by meaning
- Pairwise Ranking Losses of Click-Through Rates Prediction for Welfare Maximization in Ad Auctions (ICML 2023 · similarity 0.91)
- The (Marginal) Value of a Search Ad: An Online Causal Framework for Repeated Second-price Auctions (ICML 2026 · similarity 0.91)
- Leveraging the Hints: Adaptive Bidding in Repeated First-Price Auctions (NeurIPS 2022 · similarity 0.90)
- No-Regret Online Autobidding Algorithms in First-price Auctions (NeurIPS 2025 · similarity 0.89)
- Online Learning under Budget and ROI Constraints via Weak Adaptivity (ICML 2024 · similarity 0.88)
- Maximizing the Success Probability of Policy Allocations in Online Systems (AAAI 2024 · similarity 0.87)
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.