Explore robust optimization and generalization in machine learning through a comprehensive lecture by John Duchi from Stanford University. Delve into the historical context of robust optimization and its modern applications in machine learning, focusing on various types of robustness. Examine the potential of these approaches to certify performance levels in machine learning systems, while acknowledging their current conservative nature in practical applications. Gain insights into the connections between robust optimization and generalization, and understand their implications for developing more reliable and robust machine learning models.
Overview
Syllabus
Robust Optimization and Generalization
Taught by
Simons Institute