Optimal Real-Time Bidding Strategies

Fernandez-Tapia, Joaquin; Guéant, Olivier

doi:10.1093/amrx/abw007

Cited by 15 publications

(17 citation statements)

References 23 publications

(51 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A lot of papers study the problem of optimizing an advertiser bidding strategy in a RTB market, where the advertiser wants to maximize the number of ads it buys as well as a set of performance goals while keeping its spend below a certain budget (see [32] and references therein). [15] and [22] state the problem as a control problem and derive methods to optimize the bid online. In [34], the authors define a functional form for the bid (as a function of the impression characteristics) and write the problem as a constrained optimization problem.…”

Section: Related Workmentioning

confidence: 99%

Optimization of a SSP's Header Bidding Strategy using Thompson Sampling

Jauvion¹,

Grislain²,

Sielenou

et al. 2018

Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery &Amp; Data Mining

View full text Add to dashboard Cite

Over the last decade, digital media (web or app publishers) generalized the use of real time ad auctions to sell their ad spaces. Multiple auction platforms, also called Supply-Side Platforms (SSP), were created. Because of this multiplicity, publishers started to create competition between SSPs. In this setting, there are two successive auctions: a second price auction in each SSP and a secondary, first price auction, called header bidding auction, between SSPs.In this paper, we consider an SSP competing with other SSPs for ad spaces. The SSP acts as an intermediary between an advertiser wanting to buy ad spaces and a web publisher wanting to sell its ad spaces, and needs to define a bidding strategy to be able to deliver to the advertisers as many ads as possible while spending as little as possible. The revenue optimization of this SSP can be written as a contextual bandit problem, where the context consists of the information available about the ad opportunity, such as properties of the internet user or of the ad placement.Using classical multi-armed bandit strategies (such as the original versions of UCB and EXP3) is inefficient in this setting and yields a low convergence speed, as the arms are very correlated. In this paper we design and experiment a version of the Thompson Sampling algorithm that easily takes this correlation into account. We combine this bayesian algorithm with a particle filter, which permits to handle non-stationarity by sequentially estimating the distribution of the highest bid to beat in order to win an auction. We apply this methodology on two real auction datasets, and show that it significantly outperforms more classical approaches.The strategy defined in this paper is being developed to be deployed on thousands of publishers worldwide.

show abstract

Section: Related Workmentioning

confidence: 99%

Optimization of a SSP's Header Bidding Strategy using Thompson Sampling

Jauvion¹,

Grislain²,

Sielenou

et al. 2018

Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery &Amp; Data Mining

View full text Add to dashboard Cite

show abstract

“…Modelling the problem auction by auction, makes the proposed methodology slightly impractical. [Fernandez-Tapia et al 2016] gives a very rigorous statement of the problem and solves it in cases where impressions are generated by homogeneous Poisson processes and market prices are independent and identically distributed (IID).…”

Section: Related Papersmentioning

confidence: 99%

Recurrent Neural Networks for Stochastic Control in Real-Time Bidding

Grislain¹,

Perrin²,

Thabault³

2019

Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery &Amp; Data Mining

View full text Add to dashboard Cite

Bidding in real-time auctions can be a difficult stochastic control task; especially if underdelivery incurs strong penalties and the market is very uncertain.Most current works and implementations focus on optimally delivering a campaign given a reasonable forecast of the market. Practical implementations have a feedback loop to adjust and be robust to forecasting errors, but no implementation, to the best of our knowledge, uses a model of market risk and actively anticipates market shifts.Solving such stochastic control problems in practice is actually very challenging. This paper proposes an approximate solution based on a Recurrent Neural Network (RNN) architecture that is both effective and practical for implementation in a production environment. The RNN bidder provisions everything it needs to avoid missing its goal. It also deliberately falls short of its goal when buying the missing impressions would cost more than the penalty for not reaching it.

show abstract

“…Once all bids are processed in the system, the bidder with the highest bidding price gets the ad slot and then pays the price depending on the auction type. The entire process, from the user entering the website to displaying the banner, consumes around 100 milliseconds [2]. Google explains that the first layer of the "Pipe" which is the RTB API, is available for use, but needs the second layer which is the "Brain" which is the bidder to connect to the "Pipe".…”

Section: How Does It Work?mentioning

confidence: 99%

“…RTB basically is marketers bidding on empty advertisement slots on real-time basis to achieve efficient marketing. Fernandez-Tapia et al [2] emphasize that this defines a new model in the way digital inventory is purchased: advertisers can buy online inventory through real-time auctions for displaying a banner (or a short video). This model enables advertisers to target individual customers based on their website surfing by using browser cookie information.…”

Section: Introductionmentioning

confidence: 99%

Programmatic Advertisement and Real Time Bidding Utilization

AlSabeeh

Moghrabi

2017

Digital Nations – Smart Cities, Innovation, and Sustainability

View full text Add to dashboard Cite

This paper focuses on the use of Real Time Bidding in Programmatic Advertisement and its utilization with a focus on Kuwait. Real Time Bidding is a new marketing technique that uses information collected from web users and exploits this information in targeting them. This technique, though it saves both time and cost, is neither well known nor utilized in Kuwait. The paper sheds light on some research done on Real Time Bidding and its applications. A two-step study is presented here that aims at testing the efficiency of Real Time Bidding. The first step evaluates the relevance of the used advertisements. The second step uses a survey to assess if the technique may prove effective in Kuwait. Our findings indicate that Real Time Bidding is the closest thing to achieve efficient marketing, yet in order for it to work in Kuwait the technology infrastructure needs to be updated and upgraded in order to achieve the benefits of Programmatic Advertisement.

show abstract

Optimal Real-Time Bidding Strategies

Cited by 15 publications

References 23 publications

Optimization of a SSP's Header Bidding Strategy using Thompson Sampling

Optimization of a SSP's Header Bidding Strategy using Thompson Sampling

Recurrent Neural Networks for Stochastic Control in Real-Time Bidding

Programmatic Advertisement and Real Time Bidding Utilization

Contact Info

Product

Resources

About