Reinforcement Learning for Email Optimization

At LeoVegas, the nature of our campaigns and promotions poses interesting challenges for email optimization. The immense level of personalisation available, from timing and frequency of sendouts to the kinds of promotional offers given, makes deep neural networks a necessary tool in our predictive modelling. In this session I’ll be presenting how our approach allows us to determine optimal characteristics on a user-by-user basis, opening the door for emails and promotions personalised on the individual level.

• Email optimization should be done in a data-driven way, ideally using reinforcement learning
• The format of promotions in the gambling industry allow for high levels of personalization in each email
• Bayesian-based optimization methods achieve the important exploration/exploitation balance across all possible emails

————————————————————————————————————————————————————

Keenan Lyon – Data Scientist | LeoVegas Group

Keenan works in the Data Science Team at LeoVegas Group in Stockholm, Sweden. Before joining, he completed his PhD in Applied Mathematics at the University of Waterloo and a Postdoctoral Fellowship at Uppsala University. In his current role he is focussed on solving business-centric problems using deep neural networks, reinforcement learning, and other advanced machine learning tools.