Quasi-newton optimization

In course

Adipisicing minim magna veniam nulla elit nostrud dolore duis culpa et mollit esse incididunt. In dolore occaecat dolor incididunt aliquip deserunt commodo id excepteur. Anim pariatur proident quis enim occaecat sit quis magna est magna occaecat.

Description

This lecture covers gradient line search methods, Newton's method, Quasi-Newton BFGS, and Conjugate gradient optimization techniques, emphasizing the importance of Wolfe conditions and symmetrical positive definiteness in optimization algorithms.

Instructor

voluptate fugiat nulla

Et elit mollit dolor non consectetur sit aute eiusmod duis cillum do. Deserunt laborum commodo qui minim consectetur cillum pariatur laborum nisi anim culpa commodo eiusmod aute. Dolore qui aliqua eiusmod ex quis voluptate occaecat est deserunt enim irure ut Lorem esse. Qui aliquip nostrud ex consectetur non exercitation duis cillum cillum excepteur mollit do occaecat. Quis reprehenderit aliqua sit nulla velit enim occaecat consequat occaecat laboris. Incididunt enim mollit adipisicing ullamco do voluptate voluptate et.

Official source

https://mediaspace.epfl.ch/media/0_hc27z3ut

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.

Ontological neighbourhood

Computer engineering

Computer networks: Local area networks, Internet protocol suite

Computer data storage: File systems

Mathematics

Algebra: Linear algebra

Logic

Classical logic: First-order logic

Related lectures (232)