HANSO: Hybrid Algorithm for Non-Smooth Optimization

A MATLAB package based on the BFGS and gradient sampling methods. For general unconstrained minimization: convex or nonconvex, smooth or nonsmooth, including BFGS, limited memory BFGS and gradient sampling methods.

HANSO is based upon work supported by the National Science Foundation (NSF). Any opinions, findings and conclusions or recommendations expressed in this material are those of the authors and do not necessarily reflect the views of NSF.

HANSO version 3.0 (including a new implementation of the Gradient Sampling algorithm with stronger convergence guarantees)

Simple examples showing how to use HANSO 3.0

More interesting nonsmooth function test suite for HANSO 3.0

Older versions of HANSO