Linear matrix pencils and noncommutative convexity

Jurij Volčič

2024-07-11

Abstract:Hermitian linear matrix pencils are ubiquitous in control theory, operator systems, semidefinite optimization, and real algebraic geometry. This survey reviews the fundamental features of the matricial solution set of a linear matrix inequality, the free spectrahedron, from the perspective of free real algebraic geometry. Namely, among matricial solution sets of noncommutative polynomial inequalities, free spectrahedra are precisely the convex ones. Furthermore, a procedure for detecting free spectrahedra and producing their representing linear matrix pencils is discussed. Finally, free spectrahedra admit a perfect Positivstellensatz, leading to a semidefinite programming formulation of eigenvalue optimization over convex matricial sets constrained by noncommutative polynomial inequalities.

Functional Analysis

What problem does this paper attempt to address?

This paper aims to explore the relationship between linear matrix pencils and non - commutative convexity. Specifically, the paper mainly focuses on the following aspects:

1. **Properties of Free Spectrahedra**:
- A free spectrahedron refers to the set of matrix solutions defined by the linear matrix inequality \( L(X)\succeq0 \), where \( L \) is a Hermitian linear matrix pencil.
- The paper discusses the special position of free spectrahedra in the set of solutions of non - commutative polynomial inequalities, especially that they are convex.

2. **Methods for Detecting Free Spectrahedra**:
- An algorithm is proposed to detect whether the solution set \( D_f \) of a given non - commutative polynomial \( f \) is a free spectrahedron and generate its corresponding linear matrix pencil representation.
- This method is based on the analysis of matrix pencils in block triangular form and their determinant varieties.

3. **The Relationship between Convexity and Matrix Convexity**:
- It is proved that in the free semi - algebraic set defined by non - commutative polynomials, convexity and matrix convexity are equivalent.
- Through the theorem of Helton and McCullough, it is shown that the free spectrahedron is the only convex free semi - algebraic set.

4. **Convex Positivstellensatz**:
- The effective sums - of - squares certificates for the semi - positive definiteness of non - commutative polynomials on free spectrahedra are discussed.
- This result is called convex Positivstellensatz, which provides a powerful tool for optimizing the eigenvalues of non - commutative polynomials.

### Problems the Paper Tries to Solve

The main problems that the paper tries to solve are:

- **How to Detect and Identify Free Spectrahedra**: An algorithm is proposed to judge whether the solution set of a given non - commutative polynomial is a free spectrahedron and generate its linear matrix pencil representation.
- **The Relationship between Convexity and Matrix Convexity**: It is proved that in the free semi - algebraic set defined by non - commutative polynomials, convexity and matrix convexity are equivalent, and the free spectrahedron is the only convex free semi - algebraic set.
- **The Semi - Positive Definiteness of Non - commutative Polynomials on Free Spectrahedra**: The effective sums - of - squares certificates for the semi - positive definiteness of non - commutative polynomials on free spectrahedra are provided, that is, convex Positivstellensatz.

These studies not only deepen the understanding of linear matrix pencils and non - commutative convexity, but also provide theoretical basis and technical support for applications in related fields.