Blacklisting the Blacklist in Online Advertising
Author:
Yeming Shi, Dstillery
Abstract:
Every day, billions of online advertising slots are bought and sold through real time bidding (RTB). In RTB, publishers sometimes reject bids to deliver ads (impressions) for some brands, due to, for example, direct deals with other brands. Publishers rarely disclose which brands they blacklist to ad buyers. Buyers bidding for a blacklisted brand waste computing resources in a low latency environment and lose an opportunity to show a good ad for a different brand. Here we describe a dynamic system developed at Dstillery that detects these (publisher, brand) combinations based on ad auction win rates and limits bidding for them to the minimum. This system demonstrates 1) a significant increase in the win rates of our bids, 2) a sizable reduction of system load, and 3) effectiveness in finding qualified non-blacklisted brands to replace blacklisted brands to show ads for. The system allows us to deliver more ad impressions while making fewer bids. In addition, we develop and demonstrate a methodology of choosing the optimal exploration- exploitation balance of the problem.
More on http://www.kdd.org/kdd2017/
KDD2017 Conference is published on http://videolectures.net/
Видео Blacklisting the Blacklist in Online Advertising канала KDD2017 video
Yeming Shi, Dstillery
Abstract:
Every day, billions of online advertising slots are bought and sold through real time bidding (RTB). In RTB, publishers sometimes reject bids to deliver ads (impressions) for some brands, due to, for example, direct deals with other brands. Publishers rarely disclose which brands they blacklist to ad buyers. Buyers bidding for a blacklisted brand waste computing resources in a low latency environment and lose an opportunity to show a good ad for a different brand. Here we describe a dynamic system developed at Dstillery that detects these (publisher, brand) combinations based on ad auction win rates and limits bidding for them to the minimum. This system demonstrates 1) a significant increase in the win rates of our bids, 2) a sizable reduction of system load, and 3) effectiveness in finding qualified non-blacklisted brands to replace blacklisted brands to show ads for. The system allows us to deliver more ad impressions while making fewer bids. In addition, we develop and demonstrate a methodology of choosing the optimal exploration- exploitation balance of the problem.
More on http://www.kdd.org/kdd2017/
KDD2017 Conference is published on http://videolectures.net/
Видео Blacklisting the Blacklist in Online Advertising канала KDD2017 video
Показать
Комментарии отсутствуют
Информация о видео
Другие видео канала
Designing AI at Scale to Power Everyday LifeEstimation of Recent Ancestral Origins of Individuals on a Large Scalestruc2vec: Learning Node Representations from Structural IdentityLearning to Generate Rock Descriptions from Multivariate Well Logs with Hierarchical AttentionA Local Algorithm for StructurePreserving Graph CutMulti-Aspect Streaming Tensor CompletionVisualizing Deep Learning Activations for Improved Malaria Cell ClassificationPlanning Bike Lanes based on SharingBikes' TrajectoriesTripoles: A New Class of Relationships in Time Series DataIncorporating Feedback into Tree-based Anomaly DetectionIndustrial Machine LearningWeisfeiler-Lehman Neural Machine for Link PredictionOptimal Reserve Price for Online Ads Trading Based on Inventory IdentificationMulti-Aspect Streaming Tensor CompletionDispatch with Confidence: Integration of machine learningLearning certifiably optimal rule lists for categorical dataInterpretable Predictions of Tree-based Ensembles via Actionable Feature TweakingMultitask Learning using Task ClusteringThe Co-Evolution Model for Social Network Evolving and Opinion MigrationRevisiting power-law distributions in spectra of real world networksKDD Business Lunch