EURO-Online login
- New to EURO? Create an account
- I forgot my username and/or my password.
- Help with cookies
(important for IE8 users)
2283. A Block Cubic Newton method with Greedy Rule
Invited abstract in session MB-32: Algorithmic Advances in Large Scale Nonconvex Optimization, stream Advances in large scale nonlinear optimization.
Monday, 10:30-12:00Room: 41 (building: 303A)
Authors (first author is the speaker)
| 1. | Andrea Cristofari
|
| Department of Civil Engineering and Computer Science Engineering, University of Rome "Tor Vergata" |
Abstract
In this talk, a Block Cubic Newton method is proposed for the unconstrained minimization of a non-convex function with Lipschitz continuous Hessian matrix.
At each iteration, the proposed method first selects a block of variables by means of a greedy rule, that is, on the basis of the optimality violation.
Then, the next point is obtained by computing an approximate minimizer of the cubic Newton model.
Global convergence to stationary points is proved and complexity results are given.
Finally, some numerical results on optimization problems from machine learning are presented.
Keywords
- Continuous Optimization
- Programming, Nonlinear
- Machine Learning
Status: accepted
Back to the list of papers