AI needs MAB (and A/B testing)
AI is more than just building models.
Experimentation plays a pivotal role in shaping the effectiveness of AI products. Multi-armed bandits (MAB), with its roots in reinforcement learning, offers a dynamic and statistically efficient approach for balancing exploration and exploitation, making it an essential tool for optimizing complex AI systems. This event aims to provide an in-depth exploration of MAB testing, highlighting its technical aspects and practical applications in AI.
Event host: Daliana Liu, ex-Amazon senior data scientist, host of "the data scientist show", 250k followers on Linkedin
Speaker: Sven schmit, Head of Statistics Engineering at Eppo, PhD in computational and mathematical engineering from Stanford University.
We'll cover:
What is MAB and how to use it to improve AI products
How does MAB compared to A/B testing
Best practices and common mistakes of MAB