Hyperight

How Rovio Personalizes Angry Birds Games Using Reinforcement Learning – Ignacio Amaya de la Peña, Rovio

video
play-rounded-fill

Session Outline

Rovio’s game teams leverage Beacon, our internal cloud services platform which among other things, enables them to leverage data to improve their games. Machine Learning is part of Beacon’s offering. With a few clicks games can start using Reinforcement Learning models with “Personalized rules” which aim to replace complex sets of rules and heuristics that currently are still common across all industries. Traditionally mobile games have tested different features using A/B testing to pick the best test variant. Using “Personalized rules” there is no need to select a globally optimal variant because our ML models will find the best variant for each individual player. In this talk the goal is to present this case study putting a special focus on how our MLOps methodology was critical to bridge the gap between experimentation and production.

Key Takeaways

– From a business point of view, you will learn how Rovio ML product offering provides value in game personalization use cases

– From a technical point of view, you will learn about the MLOps required to run Reinforcement Learning use cases in production (both contextual bandits and deep reinforcement learning) and the main challenges we faced

Add comment