Learning-Based Ad Auction Design with Externalities: The Framework and A Matching-Based Approach

Ningyuan Li; Yunxuan Ma; Yang Zhao; Zhijian Duan; Yurong Chen; Zhilin Zhang; Jian Xu; Bo Zheng; Xiaotie Deng

doi:10.1145/3580305.3599403

Learning-Based Ad Auction Design with Externalities: The Framework and A Matching-Based Approach

Ningyuan Li
, Yunxuan Ma
, Yang Zhao
, Zhijian Duan
, Yurong Chen
, Zhilin Zhang
, Jian Xu
, Bo Zheng
, Xiaotie Deng

Research output: Chapters, Conference Papers, Creative and Literary Works › RGC 32 - Refereed conference paper (with host publication) › peer-review

12 Citations (Scopus)

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. © 2023 ACM.

Original language	English
Title of host publication	KDD '23 - Proceedings of the 29th ACM SIGKDD Conference on Knowledge Discovery and Data Mining
Place of Publication	New York, NY
Publisher	Association for Computing Machinery
Pages	1291-1302
Number of pages	12
ISBN (Print)	9798400701030
DOIs	https://doi.org/10.1145/3580305.3599403
Publication status	Published - Aug 2023
Externally published	Yes
Event	29th ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD 2023) - Long Beach Convention & Entertainment Center, Long Beach, United States Duration: 6 Aug 2023 → 10 Aug 2023 https://kdd.org/kdd2023/

Publication series

Name	Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining
ISSN (Print)	2154-817X

Conference

Conference	29th ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD 2023)
Abbreviated title	KDD ’23
Place	United States
City	Long Beach
Period	6/08/23 → 10/08/23
Internet address	https://kdd.org/kdd2023/

Funding

This work is supported by the National Natural Science Foundation of China (Grant No. 62172012), supported by Alibaba Group through Alibaba Innovative Research Program, and supported by Peking University-Alimama Joint Laboratory of AI Innovation. We thank Wenhan Huang and Qian Wang for various helpful discussions. We thank all anonymous reviewers for their helpful feedback.

Research Keywords

externality
learning-based mechanism design
multi-slot ad auction
online advertising

Access Output

10.1145/3580305.3599403

Other Access Options

Check CityUHK Library

Permanent Link

Right click to copy link

Cite this

Li, N., Ma, Y., Zhao, Y., Duan, Z., Chen, Y., Zhang, Z., Xu, J., Zheng, B., & Deng, X. (2023). Learning-Based Ad Auction Design with Externalities: The Framework and A Matching-Based Approach. In KDD '23 - Proceedings of the 29th ACM SIGKDD Conference on Knowledge Discovery and Data Mining (pp. 1291-1302). (Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining). Association for Computing Machinery. https://doi.org/10.1145/3580305.3599403

Li, Ningyuan ; Ma, Yunxuan ; Zhao, Yang et al. / Learning-Based Ad Auction Design with Externalities : The Framework and A Matching-Based Approach. KDD '23 - Proceedings of the 29th ACM SIGKDD Conference on Knowledge Discovery and Data Mining. New York, NY : Association for Computing Machinery, 2023. pp. 1291-1302 (Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining).

@inproceedings{33d088f6dd354964a5e0b1adb25f26d5,

title = "Learning-Based Ad Auction Design with Externalities: The Framework and A Matching-Based Approach",

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. {\textcopyright} 2023 ACM.",

keywords = "externality, learning-based mechanism design, multi-slot ad auction, online advertising",

author = "Ningyuan Li and Yunxuan Ma and Yang Zhao and Zhijian Duan and Yurong Chen and Zhilin Zhang and Jian Xu and Bo Zheng and Xiaotie Deng",

year = "2023",

month = aug,

doi = "10.1145/3580305.3599403",

language = "English",

isbn = "9798400701030",

series = "Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining",

publisher = "Association for Computing Machinery",

pages = "1291--1302",

booktitle = "KDD '23 - Proceedings of the 29th ACM SIGKDD Conference on Knowledge Discovery and Data Mining",

address = "United States",

note = "29th ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD 2023), KDD {\textquoteright}23 ; Conference date: 06-08-2023 Through 10-08-2023",

url = "https://kdd.org/kdd2023/",

}

Li, N, Ma, Y, Zhao, Y, Duan, Z, Chen, Y, Zhang, Z, Xu, J, Zheng, B & Deng, X 2023, Learning-Based Ad Auction Design with Externalities: The Framework and A Matching-Based Approach. in KDD '23 - Proceedings of the 29th ACM SIGKDD Conference on Knowledge Discovery and Data Mining. Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, Association for Computing Machinery, New York, NY, pp. 1291-1302, 29th ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD 2023), Long Beach, California, United States, 6/08/23. https://doi.org/10.1145/3580305.3599403

Learning-Based Ad Auction Design with Externalities: The Framework and A Matching-Based Approach. / Li, Ningyuan; Ma, Yunxuan; Zhao, Yang et al.
KDD '23 - Proceedings of the 29th ACM SIGKDD Conference on Knowledge Discovery and Data Mining. New York, NY: Association for Computing Machinery, 2023. p. 1291-1302 (Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining).

Research output: Chapters, Conference Papers, Creative and Literary Works › RGC 32 - Refereed conference paper (with host publication) › peer-review

TY - GEN

T1 - Learning-Based Ad Auction Design with Externalities

T2 - 29th ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD 2023)

AU - Li, Ningyuan

AU - Ma, Yunxuan

AU - Zhao, Yang

AU - Duan, Zhijian

AU - Chen, Yurong

AU - Zhang, Zhilin

AU - Xu, Jian

AU - Zheng, Bo

AU - Deng, Xiaotie

PY - 2023/8

Y1 - 2023/8

N2 - 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. © 2023 ACM.

AB - 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. © 2023 ACM.

KW - externality

KW - learning-based mechanism design

KW - multi-slot ad auction

KW - online advertising

UR - http://www.scopus.com/inward/record.url?scp=85171385787&partnerID=8YFLogxK

UR - https://www.scopus.com/record/pubmetrics.uri?eid=2-s2.0-85171385787&origin=recordpage

U2 - 10.1145/3580305.3599403

DO - 10.1145/3580305.3599403

M3 - RGC 32 - Refereed conference paper (with host publication)

SN - 9798400701030

T3 - Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining

SP - 1291

EP - 1302

BT - KDD '23 - Proceedings of the 29th ACM SIGKDD Conference on Knowledge Discovery and Data Mining

PB - Association for Computing Machinery

CY - New York, NY

Y2 - 6 August 2023 through 10 August 2023

ER -

Li N, Ma Y, Zhao Y, Duan Z, Chen Y, Zhang Z et al. Learning-Based Ad Auction Design with Externalities: The Framework and A Matching-Based Approach. In KDD '23 - Proceedings of the 29th ACM SIGKDD Conference on Knowledge Discovery and Data Mining. New York, NY: Association for Computing Machinery. 2023. p. 1291-1302. (Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining). Epub 2023 Aug 4. doi: 10.1145/3580305.3599403

Learning-Based Ad Auction Design with Externalities: The Framework and A Matching-Based Approach

Abstract

Publication series

Conference

Funding

Research Keywords

Access Output

Other Access Options

Permanent Link

Other Files and Links

Fingerprint

Cite this