David Sweet

Experimental optimization course

• Lecture 1: Definition and motivation
• Lecture 2: A/B testing I: Overview
• Lecture 3: A/B testing II: Design
• Lecture 4: A/B testing III: Operational considerations
• Lecture 5: Evaluating and presenting results
• Lecture 7: Multi-armed bandits I: Epsilon-greedy
• Lecture 8: Multi-armed bandits II: Thompson sampling
• Lecture 9: Response surface methodology
• Lecture 10: Contextual bandits
• Lecture 11: Bayesian optimization
• Lecture 12: Bayesian optimization: Gaussian process regression
• Lecture 13: Experimenting without experiments