Skip to main content
Communication Theory Lab
Communication Theory Lab
Home
News
Events
People
All People
Principal Investigator
Research Scientists
Postdoctoral Fellows
Students
Visiting Scholars
Former Members
Former Members from Texas A&M University
Former Members from University of Minnesota
Collaborators
Alumni
Research
Publications
Teaching
Funding
Media
Contact Us
Broximal Point Method
The Broximal Point Method and Its Impact on Optimization Theory and Machine Learning
Peter Richtarik, Professor, Computer Science
Sep 21, 12:00
-
13:00
B9 R2325
Broximal Point Method
proximal algorithms
machine learning
gradient methods
This talk introduces the Broximal Point Method (BPM), a condition-number-free alternative to classical proximal methods, that replaces proximal penalties with ball constraints, surveys its theoretical extensions and distributed variants, and explains how approximate non-Euclidean Brox updates underpin practical machine learning optimizers.
