Variational Optimization
Machine Learning
2012-12-21 v2 Machine Learning
Numerical Analysis
Abstract
We discuss a general technique that can be used to form a differentiable bound on the optima of non-differentiable or discrete objective functions. We form a unified description of these methods and consider under which circumstances the bound is concave. In particular we consider two concrete applications of the method, namely sparse learning and support vector classification.
Cite
@article{arxiv.1212.4507,
title = {Variational Optimization},
author = {Joe Staines and David Barber},
journal= {arXiv preprint arXiv:1212.4507},
year = {2012}
}