HIGHLIGHTS
- who: Mahesh Chandra Mukkamala from the University of Tübingen, Tübingen, Germany have published the research: Global convergence of model function based Bregman proximal minimization algorithms, in the Journal: (JOURNAL)
- what: Based on the proposed MAP property the authors propose a globally convergent algorithm called Model BPG that unifies several existing algorithms. The authors design an abstract framework for provable globally convergent algorithms based on a quality measure for suitable approximation of the objective. The authors focus on those distances that are generated from so-called Legendre functions . The authors provide further details in . . .
If you want to have access to all the content you need to log in!
Thanks :)
If you don't have an account, you can create one here.