Salvato in:
| Autori principali: | , , , , , , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | https://arxiv.org/abs/2510.07760 |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866918328417648640 |
|---|---|
| author | Lv, Yiqin Mou, Zhiyu Xu, Miao Chen, Jinghao Wang, Qi Mao, Yixiu Qu, Yun Bai, Rongquan Yu, Chuan Xu, Jian Zheng, Bo Ji, Xiangyang |
| author_facet | Lv, Yiqin Mou, Zhiyu Xu, Miao Chen, Jinghao Wang, Qi Mao, Yixiu Qu, Yun Bai, Rongquan Yu, Chuan Xu, Jian Zheng, Bo Ji, Xiangyang |
| contents | Generative auto-bidding has demonstrated strong performance in online advertising, yet it often suffers from data scarcity in small-scale settings with limited advertiser participation. While cross-task data sharing is a natural remedy to mitigate this issue, naive approaches often introduce gradient bias due to distribution shifts across different tasks, and existing methods are not readily applicable to generative auto-bidding. In this paper, we propose Validation-Aligned Optimization (VAO), a principled data-sharing method that adaptively reweights cross-task data contributions based on validation performance feedback. Notably, VAO aligns training dynamics to prioritize updates that improve generalization on the target task, effectively leveraging auxiliary data and mitigating gradient bias. Building on VAO, we introduce a unified generative autobidding framework that generalizes across multiple tasks using a single model and all available task data. Extensive experiments on standard auto-bidding benchmarks validate the effectiveness of our approach. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2510_07760 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | VAO: Validation-Aligned Optimization for Cross-Task Generative Auto-Bidding Lv, Yiqin Mou, Zhiyu Xu, Miao Chen, Jinghao Wang, Qi Mao, Yixiu Qu, Yun Bai, Rongquan Yu, Chuan Xu, Jian Zheng, Bo Ji, Xiangyang Machine Learning Artificial Intelligence Generative auto-bidding has demonstrated strong performance in online advertising, yet it often suffers from data scarcity in small-scale settings with limited advertiser participation. While cross-task data sharing is a natural remedy to mitigate this issue, naive approaches often introduce gradient bias due to distribution shifts across different tasks, and existing methods are not readily applicable to generative auto-bidding. In this paper, we propose Validation-Aligned Optimization (VAO), a principled data-sharing method that adaptively reweights cross-task data contributions based on validation performance feedback. Notably, VAO aligns training dynamics to prioritize updates that improve generalization on the target task, effectively leveraging auxiliary data and mitigating gradient bias. Building on VAO, we introduce a unified generative autobidding framework that generalizes across multiple tasks using a single model and all available task data. Extensive experiments on standard auto-bidding benchmarks validate the effectiveness of our approach. |
| title | VAO: Validation-Aligned Optimization for Cross-Task Generative Auto-Bidding |
| topic | Machine Learning Artificial Intelligence |
| url | https://arxiv.org/abs/2510.07760 |