Autobidders with Budget and ROI Constraints: Efficiency, Regret, and Pacing Dynamics
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2023
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866929609753231360 |
|---|---|
| author | Lucier, Brendan Pattathil, Sarath Slivkins, Aleksandrs Zhang, Mengxiao |
| author_facet | Lucier, Brendan Pattathil, Sarath Slivkins, Aleksandrs Zhang, Mengxiao |
| contents | We study a game between autobidding algorithms that compete in an online advertising platform. Each autobidder is tasked with maximizing its advertiser's total value over multiple rounds of a repeated auction, subject to budget and return-on-investment constraints. We propose a gradient-based learning algorithm that is guaranteed to satisfy all constraints and achieves vanishing individual regret. Our algorithm uses only bandit feedback and can be used with the first- or second-price auction, as well as with any "intermediate" auction format. Our main result is that when these autobidders play against each other, the resulting expected liquid welfare over all rounds is at least half of the expected optimal liquid welfare achieved by any allocation. This holds whether or not the bidding dynamics converges to an equilibrium. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2301_13306 |
| institution | arXiv |
| publishDate | 2023 |
| record_format | arxiv |
| spellingShingle | Autobidders with Budget and ROI Constraints: Efficiency, Regret, and Pacing Dynamics Lucier, Brendan Pattathil, Sarath Slivkins, Aleksandrs Zhang, Mengxiao Computer Science and Game Theory Machine Learning We study a game between autobidding algorithms that compete in an online advertising platform. Each autobidder is tasked with maximizing its advertiser's total value over multiple rounds of a repeated auction, subject to budget and return-on-investment constraints. We propose a gradient-based learning algorithm that is guaranteed to satisfy all constraints and achieves vanishing individual regret. Our algorithm uses only bandit feedback and can be used with the first- or second-price auction, as well as with any "intermediate" auction format. Our main result is that when these autobidders play against each other, the resulting expected liquid welfare over all rounds is at least half of the expected optimal liquid welfare achieved by any allocation. This holds whether or not the bidding dynamics converges to an equilibrium. |
| title | Autobidders with Budget and ROI Constraints: Efficiency, Regret, and Pacing Dynamics |
| topic | Computer Science and Game Theory Machine Learning |
| url | https://arxiv.org/abs/2301.13306 |