A numerical method for the computation of the best constant in a Sobolev inequality involving the spacesH2(Ω) and C0(Ω) is presented. Green’s functions corresponding to the solution of Poisson problems are used to express the solution. This (kind of) non-smooth eigenvalue problem is then formulated as a constrained optimization problem and solved with two different strategies: an augmented Lagr...

A variety of learning objectives have been recently proposed for training generative models. We show that many of them, including InfoGAN, ALI/BiGAN, ALICE, CycleGAN, VAE, β-VAE, adversarial autoencoders, AVB, and InfoVAE, are Lagrangian duals of the same primal optimization problem. This generalization reveals the implicit modeling trade-offs between flexibility and computational requirements ...

This paper deals with necessary and sufficient optimality conditions for control problems governed by semilinear elliptic partial differential equations with finitely many equality and inequality state constraints. Some recent results on this topic for optimal control problems based upon results for abstract optimization problems are compared with some new results using methods adapted to the c...

This book is concerned with theory, algorithms and software for solving nonconvex mixed integer nonlinear programs. It consists of two parts. The first part describes basic optimization tools, such as block-separable reformulations, convex and Lagrangian relaxations, decomposition methods and global optimality criteria. The second part is devoted to algorithms. Starting with a short overview on...

Based on the classic augmented Lagrangian multiplier method, we propose, analyze and test an algorithm for solving a class of equality-constrained nonsmooth optimization problems (chiefly but not necessarily convex programs) with a particular structure. The algorithm effectively combines an alternating direction technique with a nonmonotone line search to minimize the augmented Lagrangian funct...

The analytic center cutting plane method (ACCPM) is one of successful methods to solve nondifferentiable optimization problems. In this paper, ACCPM is used to accelerate Lagrangian relaxation procedure for solving a vehicle routing problem with time windows (VRPTW). First, a basic cutting plane algorithm and its relationship with a column generation technique is clarified. Then, the proposed m...

Abstract This paper presents a canonical duality theory for solving a general nonconvex 1 quadratic minimization problem with nonconvex constraints. By using the canonical dual 2 transformation developed by the first author, the nonconvex primal problem can be con3 verted into a canonical dual problem with zero duality gap. A general analytical solution 4 form is obtained. Both global and local...

We consider the global and local convergence properties of a class of augmented Lagrangian methods for solving nonlinear programming problems. In these methods, linear and more general constraints are handled in different ways. The general constraints are combined with the objective function in an augmented Lagrangian. The iteration consists of solving a sequence of subproblems; in each subprob...

This paper presents an augmented Lagrangian algorithm to solve continuous constrained global optimization problems. The algorithm approximately solves a sequence of bound constrained subproblems whose objective function penalizes equality and inequality constraints violation and depends on the Lagrange multiplier vectors and a penalty parameter. Each subproblem is solved by a population-based m...

Motzkin and Straus established a remarkable connection between the maximum clique and the Lagrangian of a graph in 1965. This connection and its extensions were successfully employed in optimization to provide heuristics for the maximum clique number in graphs. It is useful in practice if similar results hold for hypergraphs. In this paper, we provide upper bounds on the Lagrangian of a hypergr...