Lecture
Riemannian gradient descent
In course
DEMO: esse ad pariatur
Deserunt magna amet tempor eiusmod irure est enim irure officia qui. Aute ad ea pariatur reprehenderit. Nisi et aute tempor laborum.
Description
This lecture covers Riemannian gradient descent, focusing on Taylor expansions, first-order optimality conditions, algorithm templates, line search, sufficient decrease, regularity conditions, and critical points. It also discusses the behavior of local minimizers and the concept of critical points. The instructor explains the pull back of functions through retractions, the definition of local minimizers, and the necessary conditions for a point to be a critical or stationary point. The lecture concludes with a theorem on the sequence produced by Riemannian gradient descent under specific assumptions.
Official source
About this result
This page is automatically generated and may contain information that is not correct, complete, up-to-date, or relevant to your search query. The same applies to every other page on this website. Please make sure to verify the information with EPFL's official sources.
Related lectures (31)
Lipschitz Gradient Theorem
Covers the Lipschitz gradient theorem and its applications in function optimization.
Constrained optimization: the basics
Covers the basics of constrained optimization, including tangent directions, trust-region subproblems, and necessary optimality conditions.
Algorithms & Growth of Functions
Covers optimization algorithms, stable matching, and Big-O notation for algorithm efficiency.
Introduction to Optimization
Introduces linear algebra, calculus, and optimization basics in Euclidean spaces, emphasizing the power of optimization as a modeling tool.
Optimization Methods: Theory Discussion
Explores optimization methods, including unconstrained problems, linear programming, and heuristic approaches.