Symmetry reduction in AM/GM-based optimization

Moustrou, Philippe; Naumann, Helen; Riener, Cordian; Theobald, Thorsten; Verdure, Hugues

doi:10.48550/arxiv.2102.12913

Cited by 3 publications

(6 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For non-polyhedral sets X, in general the number of X-circuits is not finite anymore. It remains a future task to study necessary and sufficient criteria for sublinear circuits of structured non-polyhedral sets, such as sets with symmetry; for recent work on symmetric SAGE-based optimization see [14]. In a different direction, Forsgård and de Wolff [7] have characterized the boundary of the SAGE cone through a connection between circuits and tropical geometry.…”

Section: Discussionmentioning

confidence: 99%

Sublinear circuits for polyhedral sets

Naumann¹,

Theobald²

2021

Preprint

Self Cite

View full text Add to dashboard Cite

Sublinear circuits are generalizations of the affine circuits in matroid theory, and they arise as the convex-combinatorial core underlying constrained non-negativity certificates of exponential sums and of polynomials based on the arithmetic-geometric inequality. Here, we study the polyhedral combinatorics of sublinear circuits for polyhedral constraint sets.We give results on the relation between the sublinear circuits and their supports and provide necessary as well as sufficient criteria for sublinear circuits. Based on these characterizations, we provide some explicit results and enumerations for two prominent polyhedral cases, namely the non-negative orthant and the cube [−1, 1] n .

show abstract

Section: Discussionmentioning

confidence: 99%

Sublinear circuits for polyhedral sets

Naumann¹,

Theobald²

2021

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…Conditional SAGE is a relatively new concept. Progress has been made in understanding this technique through a convex-combinatorial structural analysis of "X-SAGE cones" [51,53], group-theoretic dimension reduction techniques [44], and a Positivstellensatz for signomial nonnegativity over compact convex sets [66]. These methods also have demonstrated applications in engineering [64].…”

Section: Related Workmentioning

confidence: 99%

Algebraic perspectives on signomial optimization

Dressler¹,

Murray²

2021

Preprint

View full text Add to dashboard Cite

Signomials are obtained by generalizing polynomials to allow for arbitrary real exponents. This generalization offers great expressive power, but has historically sacrificed the organizing principle of "degree" that is central to polynomial optimization theory. We reclaim that principle here through the concept of signomial rings, which we use to derive complete convex relaxation hierarchies of upper and lower bounds for signomial optimization via sums of arithmetic-geometric exponentials (SAGE) nonnegativity certificates. The Positivstellensatz underlying the lower bounds relies on the concept of conditional SAGE and removes regularity conditions required by earlier works, such as convexity and Archimedeanity of the feasible set. Through worked examples we illustrate the practicality of this hierarchy in areas such as chemical reaction network theory and chemical engineering. These examples include comparisons to direct global solvers (e.g., BARON and ANTIGONE) and the Lasserre hierarchy (where appropriate). The completeness of our hierarchy of upper bounds follows from a generic construction whereby a Positivstellensatz for signomial nonnegativity over a compact set provides for arbitrarily strong outer approximations of the corresponding cone of nonnegative signomials. While working toward that result, we prove basic facts on the existence and uniqueness of solutions to signomial moment problems.

show abstract

“…Thus, the number of resulting equations reduces to dim (R n ) Stab( β) by projecting onto this subspace. As a conclusion, we state the following theorem from [Mou+21] without proof: Remark 4.2.9. For special groups such as the symmetric group, the relative entropy programs can be simplified even more using various combinatorial techniques.…”

Section: Symmetry Reduction In Relative Entropy Programmingmentioning

confidence: 85%

“…Chapter 3 is based on joint work with Lukas Katthän and Thorsten Theobald and is contained in [KNT21] although the results were obtained in a different setting. Chapter 4 is based on parts of [Mou+21] and also partially on [Dre+20], the former is joint work with Philippe Moustrou, Cordian Riener, Thorsten Theobald, and Hugues Verdure, the latter is joint work with Mareike Dressler, Janin Heuer, and Timo de Wolff. Chapter 5 is based on joint work with Thorsten Theobald and contained in [NT21b], and Chapter 6 is based on the two works [NT21a] and on selected parts of [MNT20], where the former is joint work with Thorsten Theobald, and the latter is joint work with Riley Murray and Thorsten Theobald.…”

mentioning

confidence: 99%

“…Chapter 5 is based on joint work with Thorsten Theobald and contained in [NT21b], and Chapter 6 is based on the two works [NT21a] and on selected parts of [MNT20], where the former is joint work with Thorsten Theobald, and the latter is joint work with Riley Murray and Thorsten Theobald. Chapter 7 is my own work, except for the results on symmetries for the X-SAGE-cone, which come from [Mou+21].…”

mentioning

confidence: 99%

See 1 more Smart Citation

AM/GM-based optimization : gheometry and generalizations

Naumann¹

Self Cite

View full text Add to dashboard Cite

The problem of unconstrained or constrained optimization occurs in many branches of mathematics and various fields of application. It is, however, an NP-hard problem in general. In this thesis, we examine an approximation approach based on the class of SAGE exponentials, which are nonnegative exponential sums. We examine this SAGE-cone, its geometry, and generalizations. The thesis consists of three main parts: 1. In the first part, we focus purely on the cone of sums of globally nonnegative exponential sums with at most one negative term, the SAGE-cone. We ex- amine the duality theory, extreme rays of the cone, and provide two efficient optimization approaches over the SAGE-cone and its dual. 2. In the second part, we introduce and study the so-called S-cone, which pro- vides a uniform framework for SAGE exponentials and SONC polynomials. In particular, we focus on second-order representations of the S-cone and its dual using extremality results from the first part. 3. In the third and last part of this thesis, we turn towards examining the con- ditional SAGE-cone. We develop a notion of sublinear circuits leading to new duality results and a partial characterization of extremality. In the case of poly- hedral constraint sets, this examination is simplified and allows us to classify sublinear circuits and extremality for some cases completely. For constraint sets with certain conditions such as sets with symmetries, conic, or polyhedral sets, various optimization and representation results from the unconstrained setting can be applied to the constrained case.

show abstract

Symmetry reduction in AM/GM-based optimization

Cited by 3 publications

References 31 publications

Sublinear circuits for polyhedral sets

Sublinear circuits for polyhedral sets

Algebraic perspectives on signomial optimization

AM/GM-based optimization : gheometry and generalizations

Contact Info

Product

Resources

About