Research map / Bandits and online learning / Online convex optimization and games
A Combinatorial-Bandit Algorithm for the Online Joint Bid/Budget Optimization of Pay-per-Click Advertising Campaigns
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 |
|---|---|---|---|
| replace human experts with AdComB-TS for bid/budget optimization | human expert management · Real-world campaign | cost per acquisition (EUR): 83 -> 56 | Helped |
| replace human experts with AdComB-TS | human expert management · Real-world campaign | acquisitions: 350 per month -> 353 over 2 months | No clear effect |
| add GP posterior-sampling exploration (TS/BUCB) instead of using mean estimates (ablation) | AdComB-Mean · Synthetic Yahoo-based, N=4 | pseudo regret at t=100: about 35% larger regret for Mean | Helped |
| use decoupled two 1-D GPs (clicks and cost) instead of a single 2-D GP (ablation) | AdComB-2D-TS · Synthetic Yahoo-based, N=4 | pseudo regret at t=100: 2D regret more than 100% larger | Helped |
| use BayesUCB instead of Thompson Sampling for exploration (ablation) | AdComB-TS · Synthetic Yahoo-based, N=4 | pseudo regret | No clear effect |
| increase number of bid values from 5 to 20 (ablation) | AdComB-TS · Synthetic | pseudo regret at T | No clear effect |
| increase number of subcampaigns from 3 to 6 (ablation) | AdComB-TS/BUCB · Synthetic | pseudo regret at T | No clear effect |
Closest papers by meaning
- Maximizing the Success Probability of Policy Allocations in Online Systems (AAAI 2024 · similarity 0.87)
- Optimizing the coalition gain in Online Auctions with Greedy Structured Bandits (NeurIPS 2024 · similarity 0.87)
- Dynamic Budget Throttling in Repeated Second-Price Auctions (AAAI 2024 · similarity 0.86)
- Optimal Auction Design in the Joint Advertising (ICML 2025 · similarity 0.86)
- Combinatorial Stochastic-Greedy Bandit (AAAI 2024 · similarity 0.86)
- No-Regret Online Autobidding Algorithms in First-price Auctions (NeurIPS 2025 · similarity 0.86)
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.