Thesis (Ph. D.)--Brown University, 1935. / Photoprinted. "Of the original version, only the fourth and fifth chapters are here reprinted in full. The second and third have been considerably abridged ... The first chapter is even more abridged."--Foreword. Bibliography: p. i-xii; "References": p. xiii-xv.

Thesis (Ph. D.)--University of Wisconsin--Madison, 1938. / Typescript. Includes abstract and vita. eContent provider-neutral record in process. Description based on print version record. Includes bibliographical references (leaves [i]-xix).

Les gestionnaires des réseaux de transport d'électricité doivent adapter leurs outils d'aide à la décision aux avancées technologiques du XXIième siècle. Une opération sous-jacente à beaucoup d'outils est de calculer les flux en actif/réactif qui minimisent les pertes ou les coûts de production. Mathématiquement, il s'agit d'un problème d'optimisation qui peut être décrit en utilisant seulement l'addition et la multiplication de nombres complexes. L'objectif de cette thèse est de trouver des solutions globales. Un des aboutissements de ce projet doctoral hautement collaboratif est d'utiliser des résultats récents en géométrie algébrique pour calculer des flux optimaux dans le réseau Européen à haute tension. / Transmission system operators need to adapt their decision-making tools to the technological evolutions of the twenty first century. A computation inherent to most tools seeks to find alternating-current power flows that minimize power loss or generation cost. Mathematically, it consists in an optimization problem that can be described using only addition and multiplication of complex numbers. The objective of this thesis is to find global solutions, in other words the best solutions to the problem. One of the outcomes of this highly collaborative doctoral project is to use recent results from algebraic geometry to compute globally optimal power flows in the European high-voltage transmission network.

Cette thèse s’intéresse au calcul de solutions globales de problèmes de commande optimaleen boucle ouverte. La méthodologie générale se base sur l’approche par les moments, oùun problème d’optimisation est relâché en un problème généralisé des moments, dont unehiérarchie de relaxations semi-définies peut être résolue numériquement. L’approche esttout d’abord appliquée aux problèmes impulsionnels linéaires à temps variant, en modélisantle contrôle par une mesure. Les conditions semi-définies qui en résultent permettentde s’affranchir complètement des difficultés liées à la discrétisation temporelle. Ensuite, ense basant sur le formalisme des mesures d’occupations, la méthode peut être étendue auxsystèmes impulsionnels non-linéaires, et fournit une suite monotone de bornes inférieuresau coût optimal. Enfin, les résultats précédents peuvent être transposés aux systèmes àcommutation, en modélisant chaque mode par une mesure d’occupation associée. Ceci permetd’obtenir des gains substantiels en charge de calcul par rapport à l’approche classiqueoù l’espace de contrôle est mesuré / This thesis details a global method for optimal control of open-loop systems. This is doneby relaxing the control problem as a generalized moment problem, which can be solvednumerically by a hierarchy of semi-definite relaxations. The approach is first applied tothe impulsive control of linear time varying systems, by modeling the controls by a measure.The resulting semi-definite conditions circumvent time discretiziation and relateddifficulties. By the use of occupation measures, the method is then extended to a classof impulsive non-linear problems. This results in a monotone sequence of lower boundsto the original control problem. Finally, those results are transposed to switched system,by modeling each mode by a corresponding occupation measure. This allows for largecomputational gains with respect to the classical approach, where the control space ismeasured

Graph partitioning is a fundamental optimization problem that has been intensively studied. Many graph partitioning formulations are important as building blocks for divide-and-conquer algorithms on graphs as well as to many applications such as VLSI layout, packet routing in distributed networks, clustering and image segmentation. Unfortunately such problems are notorious for the huge gap between known best known approximation algorithms and hardness of approximation results. In this thesis, we study approximation algorithms for graph partitioning problems using a strong hierarchy of relaxations based on semi-definite programming, called Lasserre Hierachy. Our main contribution in this thesis is a propagation based rounding framework for solutions arising from such relaxations. We present a novel connection between the quality of solutions it outputs and column based matrix reconstruction problem. As part of our work, we derive optimal bounds on the number of columns necessary together with efficient randomized and deterministic algorithms to find such columns. Using this framework, we derive approximation schemes for many graph partitioning problems with running times dependent on how fast the graph spectrum grows. Our final contribution is a fast SDP solver for this rounding framework: Even though SDP relaxation has nO(r) many variables, we achieve running times of the form 2O(r) poly(n) by only partially solving the relevant part of relaxation. In order to achieve this, we present a new ellipsoid algorithm that returns certificate of infeasibility.

The inapproximability for NP-hard combinatorial optimization problems lies in the heart of theoretical computer science. A negative result can be either conditional, where the starting point is a complexity assumption, or unconditional, where the inapproximability holds for a restricted model of computation. Algorithms based on Linear Programming (LP) and Semidefinite Programming (SDP) relaxations are among the most prominent models of computation. The related and common measure of efficiency is the integrality gap, which sets the limitations of the associated algorithmic schemes. A number of systematic procedures, known as lift-and-project systems, have been proposed to improve the integrality gap of standard relaxations. These systems build strong hierarchies of either LP relaxations, such as the Lovasz-Schrijver (LS) and the Sherali-Adams (SA) systems, or SDP relaxations, such as the Lovasz-Schrijver SDP (LS+), the Sherali-Adams SDP (SA+) and the Lasserre (La) systems. In this thesis we prove integrality gap lower bounds for the aforementioned lift-and-project systems and for a number of combinatorial optimization problems, whose inapproximability is yet unresolved. Given that lift-and-project systems produce relaxations that have given the best algorithms known for a series of combinatorial problems, the lower bounds can be read as strong evidence of the inapproximability of the corresponding optimization problems. From the results found in the thesis we highlight the following: For every epsilon>0, the level-Omega(sqrt(log n/ log log n)) LS+ relaxation of the Vertex Cover polytope has integrality gap 2-epsilon. The integrality gap of the standard SDP for Vertex Cover remains 2-o(1) even if all hypermetric inequalities are added to the relaxation. The resulting relaxations are incomparable to the SDP relaxations derived by the LS+ system. Finally, the addition of all ell1 inequalities eliminates all solutions not in the integral hull. For every epsilon>0, the level-Omega(sqrt(log n/ log log n)) SA relaxation of Vertex Cover has integrality gap 2-epsilon. The integrality gap remains tight even for superconstant-level SA+ relaxations. We prove a tight lower bound for the number of tightenings that the SA system needs in order to prove the Pigeonhole Principle. We also prove sublinear and linear rank bounds for the La and SA systems respectively for the Tseitin tautology. Linear levels of the SA+ system treat highly unsatisfiable instances of fixed predicate-P constraint satisfaction problems over q-ary alphabets as fully satisfiable, when the satisfying assignments of the predicates P can be equipped with a balanced and pairwise independent distribution. We study the performance of the Lasserre system on the cut polytope. When the input is the complete graph on 2d+1 vertices, we show that the integrality gap is at least 1+1/(4d(d+1)) for the level-d SDP relaxation.

The inapproximability for NP-hard combinatorial optimization problems lies in the heart of theoretical computer science. A negative result can be either conditional, where the starting point is a complexity assumption, or unconditional, where the inapproximability holds for a restricted model of computation. Algorithms based on Linear Programming (LP) and Semidefinite Programming (SDP) relaxations are among the most prominent models of computation. The related and common measure of efficiency is the integrality gap, which sets the limitations of the associated algorithmic schemes. A number of systematic procedures, known as lift-and-project systems, have been proposed to improve the integrality gap of standard relaxations. These systems build strong hierarchies of either LP relaxations, such as the Lovasz-Schrijver (LS) and the Sherali-Adams (SA) systems, or SDP relaxations, such as the Lovasz-Schrijver SDP (LS+), the Sherali-Adams SDP (SA+) and the Lasserre (La) systems. In this thesis we prove integrality gap lower bounds for the aforementioned lift-and-project systems and for a number of combinatorial optimization problems, whose inapproximability is yet unresolved. Given that lift-and-project systems produce relaxations that have given the best algorithms known for a series of combinatorial problems, the lower bounds can be read as strong evidence of the inapproximability of the corresponding optimization problems. From the results found in the thesis we highlight the following: For every epsilon>0, the level-Omega(sqrt(log n/ log log n)) LS+ relaxation of the Vertex Cover polytope has integrality gap 2-epsilon. The integrality gap of the standard SDP for Vertex Cover remains 2-o(1) even if all hypermetric inequalities are added to the relaxation. The resulting relaxations are incomparable to the SDP relaxations derived by the LS+ system. Finally, the addition of all ell1 inequalities eliminates all solutions not in the integral hull. For every epsilon>0, the level-Omega(sqrt(log n/ log log n)) SA relaxation of Vertex Cover has integrality gap 2-epsilon. The integrality gap remains tight even for superconstant-level SA+ relaxations. We prove a tight lower bound for the number of tightenings that the SA system needs in order to prove the Pigeonhole Principle. We also prove sublinear and linear rank bounds for the La and SA systems respectively for the Tseitin tautology. Linear levels of the SA+ system treat highly unsatisfiable instances of fixed predicate-P constraint satisfaction problems over q-ary alphabets as fully satisfiable, when the satisfying assignments of the predicates P can be equipped with a balanced and pairwise independent distribution. We study the performance of the Lasserre system on the cut polytope. When the input is the complete graph on 2d+1 vertices, we show that the integrality gap is at least 1+1/(4d(d+1)) for the level-d SDP relaxation.

Cette thèse s'intéresse au calcul de solutions globales de problèmes de commande optimale en boucle ouverte. La méthodologie générale se base sur l'approche par les moments, où un problème d'optimisation est relâché en un problème généralisé des moments, dont une hiérarchie de relaxations semi-définies peut être résolue numériquement. L'approche est tout d'abord appliquée aux problèmes impulsionnels linéaires à temps variant, en modélisant le contrôle par une mesure. Les conditions semi-définies qui en résultent permettent de s'affranchir complètement des difficultés liées à la discrétisation temporelle. Ensuite, en se basant sur le formalisme des mesures d'occupations, la méthode peut être étendue aux systèmes impulsionnels non-linéaires, et fournit une suite monotone de bornes inférieures au coût optimal. Enfin, les résultats précédents peuvent être transposés aux systèmes à commutation, en modélisant chaque mode par une mesure d'occupation associée. Ceci permet d'obtenir des gains substantiels en charge de calcul par rapport à l'approche classique où l'espace de contrôle est mesuré.