ABSTRACT
Learning-based ad auctions have increasingly been adopted in online advertising. However, existing approaches neglect externalities, such as the interaction between ads and organic items. In this paper, we propose a general framework, namely Score-Weighted VCG, for designing learning-based ad auctions that account for externalities. The framework decomposes the optimal auction design into two parts: designing a monotone score function and an allocation algorithm, which facilitates data-driven implementation. Theoretical results demonstrate that this framework produces the optimal incentive-compatible and individually rational ad auction under various externality-aware CTR models while being data-efficient and robust. Moreover, we present an approach to implement the proposed framework with a matching-based allocation algorithm. Experiment results on both real-world and synthetic data illustrate the effectiveness of the proposed approach.
Supplemental Material
- Yang Cai, Constantinos Daskalakis, and S. Matthew Weinberg. 2012. Optimal Multi-dimensional Mechanism Design: Reducing Revenue to Welfare Maximization. 2012 IEEE 53rd Annual Symposium on Foundations of Computer Science (2012), 130--139.Google ScholarDigital Library
- Ruggiero Cavallo, Maxim Sviridenko, and Christopher A. Wilkens. 2018. Matching Auctions for Search and Native Ads. Proceedings of the 2018 ACM Conference on Economics and Computation (2018).Google ScholarDigital Library
- Chi Chen, Hui Chen, Kangzhi Zhao, Junsheng Zhou, Li He, Hongbo Deng, Jian Xu, Bo Zheng, Yong Zhang, and Chunxiao Xing. 2022. EXTR: Click-Through Rate Prediction with Externalities in E-Commerce Sponsored Search. In Proceedings of the 28th ACM SIGKDD Conference on Knowledge Discovery and Data Mining. 2732--2740.Google ScholarDigital Library
- Michael J Curry, Uro Lyi, Tom Goldstein, and John P Dickerson. 2022. Learning revenue-maximizing auctions with differentiable matching. In International Conference on Artificial Intelligence and Statistics. PMLR, 6062--6073.Google Scholar
- Xiaotie Deng, Yang Sun, Ming Yin, and Yunhong Zhou. 2010. Mechanism design for multi-slot ads auction in sponsored search markets. In Frontiers in Algorithmics: 4th International Workshop, FAW 2010, Wuhan, China, August 11-13, 2010. Proceedings 4. Springer, 11--22.Google ScholarCross Ref
- Zhijian Duan, Haoran Sun, Yurong Chen, and Xiaotie Deng. 2023. A Scalable Neural Network for DSIC Affine Maximizer Auction Design. arXiv preprint arXiv:2305.12162 (2023).Google Scholar
- Zhijian Duan, Jingwu Tang, Yutong Yin, Zhe Feng, Xiang Yan, Manzil Zaheer, and Xiaotie Deng. 2022. A context-integrated transformer-based neural network for auction design. In International Conference on Machine Learning. PMLR, 5609--5626.Google Scholar
- Paul Dütting, Zhe Feng, Harikrishna Narasimhan, David Parkes, and Sai Srivatsa Ravindranath. 2019. Optimal auctions through deep learning. In International Conference on Machine Learning. PMLR, 1706--1715.Google Scholar
- Benjamin Edelman, Michael Ostrovsky, and Michael Schwarz. 2007. Internet advertising and the generalized second-price auction: Selling billions of dollars worth of keywords. American economic review, Vol. 97, 1 (2007), 242--259.Google Scholar
- Nicola Gatti, Alessandro Lazaric, and Francesco Trovò. 2012. A truthful learning mechanism for contextual multi-slot sponsored search auctions with externalities. In Proceedings of the 13th ACM Conference on Electronic Commerce. 605--622.Google ScholarDigital Library
- Shivam Gupta. 2016. On a modification of the VCG mechanism and its optimality. Oper. Res. Lett., Vol. 44 (2016), 415--418.Google ScholarDigital Library
- Dmitry Ivanov, Iskander Safiulin, Igor Filippov, and Ksenia Balabaeva. 2022. Optimal-er Auctions through Attention. In Advances in Neural Information Processing Systems.Google Scholar
- Bernard J Jansen and Tracy Mullen. 2008. Sponsored search: an overview of the concept, history, and technology. International Journal of Electronic Business, Vol. 6, 2 (2008), 114--131.Google ScholarCross Ref
- Przemyslaw Jeziorski and Ilya Segal. 2015. What makes them click: Empirical analysis of consumer demand for search advertising. American Economic Journal: Microeconomics, Vol. 7, 3 (2015), 24--53.Google ScholarCross Ref
- J. Kiefer. 1957. Optimum Sequential Search and Approximation Methods Under Minimum Regularity Assumptions. Journal of The Society for Industrial and Applied Mathematics, Vol. 5 (1957), 105--136.Google ScholarCross Ref
- Roger W. Koenker and Gilbert W. Jr. Bassett. 2007. Regression Quantiles.Google Scholar
- Harold W. Kuhn. 1955. The Hungarian method for the assignment problem. Naval Research Logistics (NRL), Vol. 52 (1955).Google ScholarCross Ref
- Guogang Liao, Xuejian Li, Ze Wang, Fan Yang, Muzhi Guan, Bingqi Zhu, Yongkang Wang, Xingxing Wang, and Dong Wang. 2022. NMA: Neural Multi-slot Auctions with Externalities for Online Advertising. arXiv preprint arXiv:2205.10018 (2022).Google Scholar
- Xiangyu Liu, Chuan Yu, Zhilin Zhang, Zhenzhe Zheng, Yu Rong, Hongtao Lv, Da Huo, Yiqing Wang, Dagui Chen, Jian Xu, et al. 2021. Neural auction: End-to-end learning of auction mechanisms for e-commerce advertising. In Proceedings of the 27th ACM SIGKDD Conference on Knowledge Discovery & Data Mining. 3354--3364.Google ScholarDigital Library
- Roger B Myerson. 1981. Optimal auction design. Mathematics of operations research, Vol. 6, 1 (1981), 58--73.Google Scholar
- Jad Rahme, Samy Jelassi, and S. Matthew Weinberg. 2021. Auction Learning as a Two-Player Game. In International Conference on Learning Representations.Google Scholar
- Tuomas Sandholm and Anton Likhodedov. 2015. Automated design of revenue-maximizing combinatorial auctions. Operations Research, Vol. 63, 5 (2015), 1000--1025.Google ScholarDigital Library
- Shai Shalev-Shwartz and Shai Ben-David. 2014. Understanding machine learning: From theory to algorithms. Cambridge university press.Google ScholarDigital Library
- Weiping Song, Chence Shi, Zhiping Xiao, Zhijian Duan, Yewen Xu, Ming Zhang, and Jian Tang. 2019. Autoint: Automatic feature interaction learning via self-attentive neural networks. In Proceedings of the 28th ACM International Conference on Information and Knowledge Management. 1161--1170.Google ScholarDigital Library
- Zhulin Tao, Xiang Wang, Xiangnan He, Xianglin Huang, and Tat-Seng Chua. 2020. HoAFM: a high-order attentive factorization machine for CTR prediction. Information Processing & Management, Vol. 57, 6 (2020), 102076.Google ScholarCross Ref
- Yiqing Wang, Junqi Jin, Zhenzhe Zheng, Haiyang Xu, Fan Wu, Yuning Jiang, and Guihai Chen. 2021. Multi-objective Dynamic Auction Mechanism for Online Advertising. In 2021 IEEE International Performance, Computing, and Communications Conference (IPCCC). IEEE, 1--8.Google Scholar
- Yiqing Wang, Xiangyu Liu, Zhenzhe Zheng, Zhilin Zhang, Miao Xu, Chuan Yu, and Fan Wu. 2022. On Designing a Two-stage Auction for Online Advertising. In Proceedings of the ACM Web Conference 2022. 90--99.Google ScholarDigital Library
- Chao Wen, Miao Xu, Zhilin Zhang, Zhenzhe Zheng, Yuhui Wang, Xiangyu Liu, Yu Rong, Dong Xie, Xiaoyang Tan, Chuan Yu, et al. 2022. A Cooperative-Competitive Multi-Agent Framework for Auto-bidding in Online Advertising. In Proceedings of the Fifteenth ACM International Conference on Web Search and Data Mining. 1129--1139.Google ScholarDigital Library
- Zhilin Zhang, Xiangyu Liu, Zhenzhe Zheng, Chenrui Zhang, Miao Xu, Junwei Pan, Chuan Yu, Fan Wu, Jian Xu, and Kun Gai. 2021. Optimizing Multiple Performance Metrics with Deep GSP Auctions for E-commerce Advertising. In Proceedings of the 14th ACM International Conference on Web Search and Data Mining. 993--1001.Google ScholarDigital Library
Index Terms
- Learning-Based Ad Auction Design with Externalities: The Framework and A Matching-Based Approach
Recommendations
Neural Auction: End-to-End Learning of Auction Mechanisms for E-Commerce Advertising
KDD '21: Proceedings of the 27th ACM SIGKDD Conference on Knowledge Discovery & Data MiningIn e-commerce advertising, it is crucial to jointly consider various performance metrics, e.g., user experience, advertiser utility, and platform revenue. Traditional auction mechanisms, such as GSP and VCG auctions, can be suboptimal due to their fixed ...
Expressive auctions for externalities in online advertising
WWW '10: Proceedings of the 19th international conference on World wide webWhen online ads are shown together, they compete for user attention and conversions, imposing negative externalities on each other. While the competition for user attention in sponsored search can be captured via models of clickthrough rates, the post-...
Mechanism design for multi-slot ads auction in sponsored search markets
FAW'10: Proceedings of the 4th international conference on Frontiers in algorithmicsIn this paper, we study pricing models for multi-slot advertisements, where advertisers can bid to place links to their sales webpages at one or multiple slots on a webpage, called the multi-slot AD auction problem. We develop and analyze several ...
Comments