Building Efficient Learning Algorithms: A Computational Regularization Perspective - Lorenzo Rosasco
Alan Turing Institute via YouTube
Overview
Explore the theoretical foundations of machine learning algorithms in this 45-minute talk by Lorenzo Rosasco at the Alan Turing Institute. Delve into the intersection of statistics, probability, and optimization to understand how algorithmic design choices impact learning outcomes. Examine a least squares learning scenario to discover how various algorithmic techniques can be unified within a regularization framework. Gain insights into building resource-efficient and accurate algorithms, moving beyond traditional empirical trial-and-error approaches. Learn how computational regularization offers a new perspective on algorithm design, potentially revolutionizing the way we approach machine learning problems.
Syllabus
Building efficient learning algorithms: a computational regularization perspective - Lorenzo Rosasco
Taught by
Alan Turing Institute