Applied Reinforcement Learning for Online Ads/Recommender


Online advertising is one of the most successful business application of Machine Learning and Data Science.

On business side, Analyst expect that Online Advertising will generate 480 billion USD of revenue by 2025. In online advertising, Machine Learning automation plays a fundamental role by providing in real time, relevant ads to billion of customers in E-commerce, streaming services, ...

In this session, we illustrate the role of machine learning, and particularly a sub-field called reinforcement learning in online advertising. Presentation will go through 3 main axes.

As a introduction, we recall the various business application of Machine Learning and Data Science in Online Advertising: from customer segmentation to online real time personalization, through the usage of Causal inference for A/B tests and uplift modelling. Particularly, we highlight the specific role of reinforcement learning in online Ads/Recommender.

Then, we focus on the foundations of Reinforcement Learning as an overall science field, explaining the technical concepts as well as the assumptions behind different approaches, including stochastic and contextual bandit algorithms. In final part, we explain how Reinforcement Learning models/frameworks are integrated in real world Ads/Recommenders systems, including description of end to end data pipelines and various real metrics measurements. Thus, we will go in depth various latency and scalability concepts and technologies used. Through this talk, we also illustrate with real life examples from the sub-field of geo-spatial advertisement.


Kevin Noel is currently Lead of Machine Learning Ads at Mapbox Japan and has more than 10 years experience in Japan. Previously, he held principal ML role at the largest Big Data, E-commerce in Japan (Rakuten), working with large scale multi-modal data (Tabular, Time series, Japanese NLP, image) through numerous machine learning projects in real time Ads/Recommendations, also provided internal training on Deep Learning and external talks on applied ML (New York, 2019, Kobe(Japan)... )... Prior to this, Kevin, with a background in applied Stochastic Modeling and Data Mining from Ecole Centrale (France), held various quantitative roles a BNP Paribas, Bank of America, and ING in Asia/Japan.

Open Data Science




Open Data Science
One Broadway
Cambridge, MA 02142

Privacy Settings
We use cookies to enhance your experience while using our website. If you are using our Services via a browser you can restrict, block or remove cookies through your web browser settings. We also use content and scripts from third parties that may use tracking technologies. You can selectively provide your consent below to allow such third party embeds. For complete information about the cookies we use, data we collect and how we process them, please check our Privacy Policy
Consent to display content from - Youtube
Consent to display content from - Vimeo
Google Maps
Consent to display content from - Google