Sum-of-squares optimization

A sum-of-squares optimization program is an optimization problem with a linear cost function and constraints that certain polynomials constructed from the decision variables should be sums of squares. When the maximum degree of the polynomials involved is fixed, sum-of-squares optimization is also known as the Lasserre hierarchy of semidefinite programming relaxations.

Sum-of-squares optimization techniques have been applied across a variety of areas, including control theory (in particular, for searching for polynomial Lyapunov functions for dynamical systems described by polynomial vector fields), statistics, finance and machine learning.