Slater Condition Research Articles

Discretization and quantification for distributionally robust optimization with decision-dependent ambiguity sets

ABSTRACT In this paper, we investigate the discrete approximation and the quantitative analysis for a class of the distributionally robust optimization problems with decision-dependent ambiguity sets. We establish the local Lipschitz continuity of the decision-dependent ambiguity set, measured by the Hausdorff distance, under a broad class of metrics known as ζ-structure and the Slater condition. Furthermore, we employ Lagrange duality and first-order growth conditions to derive quantitative analysis for the optimal value and optimal solution. We also examine the application of a classical and widely-used ambiguity set within the theoretical framework of this paper. Finally, we conduct experiments to demonstrate the computational times and variations in the optimal value.

Stochastic linear–quadratic control problems with affine constraints

This paper investigates the stochastic linear–quadratic control problems with affine constraints, in which both equality and inequality constraints are involved. With the help of the Pontryagin maximum principle and Lagrangian duality theory, both the dual problem and the state feedback form of the solution are obtained for the primal problem. Under the Slater condition, the strong duality is proved between the dual problem and the primal problem, and the KKT condition is also provided for solving the primal problem. Moreover, a new sufficient condition is given for the invertibility assumption, which ensures the uniqueness of the solutions to the dual problem. Finally, two numerical examples are provided to illustrate our main results.

A note on the generalized Hessian of the least squares associated with systems of linear inequalities

The goal of this note is to point out an erroneous formula for the generalized Hessian of the least squares associated with a system of linear inequalities, that was given in the paper ‘A finite Newton method for classification’ by Mangasarian (Optim. Methods Softw. 17 (2002), pp. 913–929) and reproduced multiple times in other publications. We also provide sufficient contiditions for the validity of Mangasarian's formula and show that Slater's condition guarantees that some particular elements from the set defined by Mangasarian belong to the generalized Hessian of the corresponding function.

Strong duality of a conic optimization problem with a single hyperplane and two cone constraints

Strong (Lagrangian) duality of general conic optimization problems (COPs) has long been studied and its profound and complicated results appear in different forms in a wide range of literatures. As a result, characterizing the known and unknown results can sometimes be difficult. The aim of this article is to provide a unified and geometric overview of strong duality of COPs for the known results, and to explain the essential ideas of the duality results with the simplest geometry. For our framework, we employ a COP minimizing a linear function in a vector variable subject to a single hyperplane constraint and two cone constraints , . It can be identically reformulated as a simpler COP with the single hyperplane constraint and the single cone constraint . This simple COP and its dual as well as their duality relation can be represented geometrically, and they have no duality gap without any constraint qualification. The dual of the original target COP is equivalent to the dual of the reformulated COP if the Minkowski sum of the duals of the two cones and is closed or if the dual of the reformulated COP satisfies a certain Slater condition. Thus, these two conditions make it possible to transfer all duality results, including the existence and/or boundedness of optimal solutions, on the reformulated COP to the ones on the original target COP, and further to the ones on a standard primal-dual pair of COPs with symmetry.

Gradient-Variation Bound for Online Convex Optimization with Constraints

We study online convex optimization with constraints consisting of multiple functional constraints and a relatively simple constraint set, such as a Euclidean ball. As enforcing the constraints at each time step through projections is computationally challenging in general, we allow decisions to violate the functional constraints but aim to achieve a low regret and cumulative violation of the constraints over a horizon of T time steps. First-order methods achieve an O(sqrt{T}) regret and an O(1) constraint violation, which is the best-known bound under the Slater's condition, but do not take into account the structural information of the problem. Furthermore, the existing algorithms and analysis are limited to Euclidean space. In this paper, we provide an instance-dependent bound for online convex optimization with complex constraints obtained by a novel online primal-dual mirror-prox algorithm. Our instance-dependent regret is quantified by the total gradient variation V_*(T) in the sequence of loss functions. The proposed algorithm works in general normed spaces and simultaneously achieves an O(sqrt{V_*(T)}) regret and an O(1) constraint violation, which is never worse than the best-known (O(sqrt{T}), O(1)) result and improves over previous works that applied mirror-prox-type algorithms for this problem achieving O(T^{2/3}) regret and constraint violation. Finally, our algorithm is computationally efficient, as it only performs mirror descent steps in each iteration instead of solving a general Lagrangian minimization problem.

Duality for Sets of Strong Slater Points

The strong Slater condition plays a significant role in the stability analysis of linear semi-infinite inequality systems. This piece of work studies the set of strong Slater points, whose non-emptiness guarantees the fullfilment of the strong Slater condition. Given a linear inequality system, we firstly establish some basic properties of the set of strong Slater points. Then, we derive dual characterizations for this set in terms of the data of the system, following similar characterizations provided also for the set of Slater points and the solution set of the given system, which are based on the polarity operators for evenly convex and closed convex sets. Finally, we present two geometric interpretations and apply our results to analyze the strict inequality systems defined by lower semicontinuous convex functions.

Strong solvability of restricted interval systems and its applications in quadratic and geometric programming

We consider interval systems of linear equations and inequalities with a restriction to some a priori given set. We focus on a characterization of strong solvability, that is, solvability for each realization of interval values, and we compare this with an existence of a strong solution defined analogously. The motivation comes from the area of interval-valued optimization problems, where strong solvability means guaranteed feasibility of any realization of the problem. Strong solvability with strict inequalities implies the robust Slater condition, which ensures that standard optimality conditions can be used. We apply the issues particularly in two optimization classes, convex quadratic programming with quadratic constraints and posynomial geometric programming. For the former, we also utilize the presented result to improve a characterization of the worst case optimal value. Eventually, we state several open problems that emerged while deriving the results.

Regularization algorithms for linear copositive problems

The paper is devoted to the regularization of linear Copositive Programming problems which consists of transforming a problem to an equivalent form, where the Slater condition is satisfied and therefore the strong duality holds. We describe regularization algorithms based on a concept of immobile indices and on the understanding of the important role that these indices play in the feasible sets' characterization. These algorithms are compared to some regularization procedures developed for a more general case of convex problems and based on a facial reduction approach. We show that the immobile-index-based approach combined with the specifics of copositive problems allows us to construct more explicit and detailed regularization algorithms for linear Copositive Programming problems than those already available.

Duality for convex infinite optimization on linear spaces

AbstractThis note establishes a limiting formula for the conic Lagrangian dual of a convex infinite optimization problem, correcting the classical version of Karney [Math. Programming 27 (1983) 75-82] for convex semi-infinite programs. A reformulation of the convex infinite optimization problem with a single constraint leads to a limiting formula for the corresponding Lagrangian dual, called sup-dual, and also for the primal problem in the case when strong Slater condition holds, which also entails strong sup-duality.

Slater Condition for Tangent Derivatives

Noting that the existing Slater condition, as a fundamental constraint qualification in optimization, is only applicable in the convex setting, we introduce and study the Slater condition for the Bouligand and Clarke tangent derivatives of a general vector-valued function F with respect to a closed convex cone K. Without any assumption, it is proved that the Slater condition for the Clarke (respectively, Bouligand) tangent derivative with respect to K is always stable when the objective function F undergoes small Lipschitz (calm) perturbations. Based on this, we prove that if the Clarke (Bouligand) tangent derivative of F satisfies the Slater condition (with respect to K) then the conic inequality determined by F has a stable metric subregularity when F undergoes small Lipschitz (calm regular) perturbations. In the composite-convexity case, the converse implication is also proved to be true. Moreover, under the Slater condition for the tangent derivative of F, it is proved that the normal cone to the sublevel set of F can be formulated by the subdifferential of F, which improves the corresponding results in either the smooth or convex case. As applications, without any qualification assumption, we improve and generalize formulas for the normal cone to a convex sublevel set by Cabot and Thibault [(2014), Sequential formulae for the normal cone to sublevel sets. Transactions of the American Mathematical Society 366(12):6591–6628]. With the help of these formulas, some new Karush–Kuhn–Tucker optimality conditions are established.

Explicit model predictive control for PDEs: The case of a heat equation

Explicit model predictive control design is carefully developed for discrete-time linear plants on Hilbert spaces, and we highlight the role of the so-called Slater condition in the reliable explicit solution of the MPC optimization. We then proceed to present an explicit MPC algorithm that accounts for the stabilization and input constraints satisfaction. We do structure preserving temporal discretization of the infinite-dimensional parabolic PDE system by application of the Cayley transformation. The salient feature of explicit MPC design is the realization of the region-free approach in explicit MPC design with identification of active constraint sets to realize optimal stabilization and constraints satisfaction. Finally, the resulting design is illustrated by the application to the PDE model given by an unstable heat equation with boundary actuation and Neumann boundary conditions. The example demonstrates simultaneous stabilization and input constraints satisfaction on the one hand, and on the ability to deal with a relatively high plant dimension and a long optimization horizon on the other hand.

Simultaneously achieving sublinear regret and constraint violations for online convex optimization with time-varying constraints

In this paper, we develop a novel virtual-queue-based online algorithm for online convex optimization (OCO) problems with long-term and time-varying constraints and conduct a performance analysis with respect to the dynamic regret and constraint violations. We design a new update rule of dual variables and a new way of incorporating time-varying constraint functions into the dual variables. To the best of our knowledge, our algorithm is the first parameter-free algorithm to simultaneously achieve sublinear dynamic regret and constraint violations. Our proposed algorithm also outperforms the state-of-the-art results in many aspects, e.g., our algorithm does not require the Slater condition. Meanwhile, for a group of practical and widely-studied constrained OCO problems in which the variation of consecutive constraints is smooth enough across time, our algorithm achieves O(1) constraint violations. Furthermore, we extend our algorithm and analysis to the case when the time horizon T is unknown. Finally, numerical experiments are conducted to validate the theoretical guarantees of our algorithm, and some applications of our proposed framework will be outlined.

Dynamic online convex optimization with long-term constraints via virtual queue

In this paper, we investigate the online convex optimization (OCO) with long-term constraints which is widely used in various resource allocations and recommendation systems. Different from the most existing works, our work adopts a dynamic benchmark to analyze the optimization performance since the dynamic benchmark is more common than the static benchmark in practical applications. Moreover, compared with many constrained OCO works ignoring the Slater condition, we study the effect of the Slater condition on the constraint violation bounds and obtain the better performance of the constraint violations when the Slater condition holds. More importantly, we propose a novel iterative optimization algorithm based on the virtual queues to achieve sublinear regret and constraint violations. Finally, we apply our dynamic OCO model to a resource allocation problem in cloud computing and the results of the experiments validate the effectiveness of our algorithm.

Solving SDP completely with an interior point oracle

We suppose the existence of an oracle which solves any semidefinite programming (SDP) problem satisfying strong feasibility (i.e. Slater's condition) simultaneously at its primal and dual sides. We note that such an oracle might not be able to directly solve general SDPs even after certain regularization schemes are applied. In this work we fill this gap and show how to use such an oracle to ‘completely solve’ an arbitrary SDP. Completely solving entails, for example, distinguishing between weak/strong feasibility/infeasibility and detecting when the optimal value is attained or not. We will employ several tools, including a variant of facial reduction where all auxiliary problems are ensured to satisfy strong feasibility at all sides. Our main technical innovation, however, is an analysis of double facial reduction, which is the process of applying facial reduction twice: first to the original problem and then once more to the dual of the regularized problem obtained during the first run. Although our discussion is focused on semidefinite programming, the majority of the results are proved for general convex cones.

Linear semidefinite programming problems: regularisation and strong dual formulations

Regularisation consists in reducing a given optimisation problem to an equivalent form where certain regularity conditions, which guarantee the strong duality, are fulfilled. In this paper, for linear problems of semidefinite programming (SDP), we propose a regularisation procedure which is based on the concept of an immobile index set and its properties. This procedure is described in the form of a finite algorithm which converts any linear semidefinite problem to a form that satisfies the Slater condition. Using the properties of the immobile indices and the described regularization procedure, we obtained new dual SDP problems in implicit and explicit forms. It is proven that for the constructed dual problems and the original problem the strong duality property holds true.

Online Primal-Dual Mirror Descent under Stochastic Constraints

We consider online convex optimization with stochastic constraints where the objective functions are arbitrarily time-varying and the constraint functions are independent and identically distributed (i.i.d.) over time. Both the objective and constraint functions are revealed after the decision is made at each time slot. The best known expected regret for solving such a problem is O( √ T ), with a coefficient that is polynomial in the dimension of the decision variable and relies on the Slater condition (i.e. the existence of interior point assumption), which is restrictive and in particular precludes treating equality constraints. In this paper, we show that such Slater condition is in fact not needed. We propose a new primal-dual mirror descent algorithm and show that one can attain O( √ T ) regret and constraint violation under a much weaker Lagrange multiplier assumption, allowing general equality constraints and significantly relaxing the previous Slater conditions. Along the way, for the case where decisions are contained in a probability simplex, we reduce the coefficient to have only a logarithmic dependence on the decision variable dimension. Such a dependence has long been known in the literature on mirror descent but seems new in this new constrained online learning scenario. Simulation experiments on a data center server provision problem with real electricity price traces further demonstrate the performance of our proposed algorithm.

Online Primal-Dual Mirror Descent under Stochastic Constraints

We consider online convex optimization with stochastic constraints where the objective functions are arbitrarily time-varying and the constraint functions are independent and identically distributed (i.i.d.) over time. Both the objective and constraint functions are revealed after the decision is made at each time slot. The best known expected regret for solving such a problem is $\mathcalO (\sqrtT )$, with a coefficient that is polynomial in the dimension of the decision variable and relies on theSlater condition (i.e. the existence of interior point assumption), which is restrictive and in particular precludes treating equality constraints. In this paper, we show that such Slater condition is in fact not needed. We propose a newprimal-dual mirror descent algorithm and show that one can attain $\mathcalO (\sqrtT )$ regret and constraint violation under a much weaker Lagrange multiplier assumption, allowing general equality constraints and significantly relaxing the previous Slater conditions. Along the way, for the case where decisions are contained in a probability simplex, we reduce the coefficient to have only a logarithmic dependence on the decision variable dimension. Such a dependence has long been known in the literature on mirror descent but seems new in this new constrained online learning scenario. Simulation experiments on a data center server provision problem with real electricity price traces further demonstrate the performance of our proposed algorithm.

Modal dynamic residual-based model updating through regularized semidefinite programming with facial reduction

Structural model updating techniques optimize model parameter values for improving the predication accuracy of a numerical model. Formulated as optimization problems, the objective of these techniques is to minimize the difference between model prediction and the experimental data. Most of these optimization problems are nonconvex, and consequently, the global optimum is very hard to solve. The sum of squares (SOS) approach and its sparse variant have been reported to relax a nonconvex polynomial optimization problem into a convex semidefinite programming (SDP) problem, which is more readily solvable. However, the corresponding convex SDP problem may fail the Slater condition qualification. The failure increases the difficulty for numerical algorithms, e.g. the interior point method, to solve the SDP problem. This paper proposes to utilize the facial reduction technique to regularize an SDP problem which fails the Slater condition qualification. In addition, the regularized SDP problem has smaller size, which helps to improve the computation efficiency. The performance of the proposed model updating approach is evaluated through a plane truss example.

Duality Theorems for Convex and Quasiconvex Set Functions

In mathematical programming, duality theorems play a central role. Especially, in convex and quasiconvex programming, Lagrange duality and surrogate duality have been studied extensively. Additionally, constraint qualifications are essential ingredients of the powerful duality theory. The best-known constraint qualifications are the interior point conditions, also known as the Slater-type constraint qualifications. A typical example of mathematical programming is a minimization problem of a real-valued function on a vector space. This types of problems have been studied widely and have been generalized in several directions. Recently, the authors investigate set functions and Fenchel duality. However, duality theorems and its constraint qualifications for mathematical programming with set functions have not been studied yet. It is expected to study set functions and duality theorems. In this paper, we study duality theorems for convex and quasiconvex set functions. We show Lagrange duality theorem for convex set functions and surrogate duality theorem for quasiconvex set functions under the Slater condition. As an application, we investigate an uncertain problem with motion uncertainty.

Exact penalty functions for optimal control problems II: Exact penalization of terminal and pointwise state constraints

SummaryThe second part of our study is devoted to an analysis of the exactness of penalty functions for optimal control problems with terminal and pointwise state constraints. We demonstrate that with the use of the exact penalty function method one can reduce fixed‐endpoint problems for linear time‐varying systems and linear evolution equations with convex constraints on the control inputs to completely equivalent free‐endpoint optimal control problems, if the terminal state belongs to the relative interior of the reachable set. In the nonlinear case, we prove that a local reduction of fixed‐endpoint and variable‐endpoint problems to equivalent free‐endpoint ones is possible under the assumption that the linearized system is completely controllable, and point out some general properties of nonlinear systems under which a global reduction to equivalent free‐endpoint problems can be achieved. In the case of problems with pointwise state inequality constraints, we prove that such problems for linear time‐varying systems and linear evolution equations with convex state constraints can be reduced to equivalent problems without state constraints, provided one uses the L∞ penalty term, and Slater's condition holds true, while for nonlinear systems a local reduction is possible, if a natural constraint qualification is satisfied. Finally, we show that the exact Lp‐penalization of state constraints with finite p is possible for convex problems, if Lagrange multipliers corresponding to the state constraints belong to Lp′, where p′ is the conjugate exponent of p, and for general nonlinear problems, if the cost functional does not depend on the control inputs explicitly.