All issues

The Lipschitz continuous property has been used for a long time to solve the global optimization problem and continues to be used. Here we can mention the work of Piyavskii, Yevtushenko, Strongin, Shubert, Sergeyev, Kvasov and others. Most papers assume a priori knowledge of the Lipschitz constant, but the derivation of this constant is a separate problem. Further still, we must prove that an objective function is really Lipschitz, and it is a complicated problem too. In the case where the Lipschitz continuity is established, Strongin proposed an algorithm for global optimization of a satisfying Lipschitz condition on a compact interval function without any a priori knowledge of the Lipschitz estimate. The algorithm not only finds a global extremum, but it determines the Lipschitz estimate too. It is known that every function that satisfies the Lipchitz condition on a compact convex set is uniformly continuous, but the reverse is not always true. However, there exist models (Arutyunova, Dulliev, Zabotin) whose study requires a minimization of the continuous but definitely not Lipschitz function. One of the algorithms for solving such a problem was proposed by R. J. Vanderbei. In his work he introduced some generalization of the Lipchitz property named $\varepsilon$-Lipchitz and proved that a function defined on a compact convex set is uniformly continuous if and only if it satisfies the $\varepsilon$-Lipchitz condition. The above-mentioned property allowed him to extend Piyavskii’s method. However, Vanderbei assumed that for a given value of $\varepsilon$ it is possible to obtain an associate Lipschitz $\varepsilon$-constant, which is a very difficult problem. Thus, there is a need to construct, for a function continuous on a compact convex domain, a global optimization algorithm which works in some way like Strongin’s algorithm, i.e., without any a priori knowledge of the Lipschitz $\varepsilon$-constant. In this paper we propose an extension of Strongin’s global optimization algorithm to a function continuous on a compact interval using the $\varepsilon$-Lipchitz conception, prove its convergence and solve some numerical examples using the software that implements the developed method.

In this article we consider min-min type of problems or minimization by two groups of variables. In some way it is similar to classic min-max saddle point problem. Although, saddle point problems are usually more difficult in some way. Min-min problems may occur in case if some groups of variables in convex optimization have different dimensions or if these groups have different domains. Such problem structure gives us an ability to split the main task to subproblems, and allows to tackle it with mixed oracles. However existing articles on this topic cover only zeroth and first order oracles, in our work we consider high-order tensor methods to solve inner problem and fast gradient method to solve outer problem.

We assume, that outer problem is constrained to some convex compact set, and for the inner problem we consider both unconstrained case and being constrained to some convex compact set. By definition, tensor methods use high-order derivatives, so the time per single iteration of the method depends a lot on the dimensionality of the problem it solves. Therefore, we suggest, that the dimension of the inner problem variable is not greater than 1000. Additionally, we need some specific assumptions to be able to use mixed oracles. Firstly, we assume, that the objective is convex in both groups of variables and its gradient by both variables is Lipschitz continuous. Secondly, we assume the inner problem is strongly convex and its gradient is Lipschitz continuous. Also, since we are going to use tensor methods for inner problem, we need it to be p-th order Lipschitz continuous ($p > 1$). Finally, we assume strong convexity of the outer problem to be able to use fast gradient method for strongly convex functions.

We need to emphasize, that we use superfast tensor method to tackle inner subproblem in unconstrained case. And when we solve inner problem on compact set, we use accelerated high-order composite proximal method.

Additionally, in the end of the article we compare the theoretical complexity of obtained methods with regular gradient method, which solves the mentioned problem as regular convex optimization problem and doesn’t take into account its structure (Remarks 1 and 2).

The work is devoted to the study of subgradient methods with different variations of the Polyak stepsize for minimization functions from the class of weakly convex and relatively weakly convex functions that have the corresponding analogue of a sharp minimum. It turns out that, under certain assumptions about the starting point, such an approach can make it possible to justify the convergence of the subgradient method with the speed of a geometric progression. For the subgradient method with the Polyak stepsize, a refined estimate for the rate of convergence is proved for minimization problems for weakly convex functions with a sharp minimum. The feature of this estimate is an additional consideration of the decrease of the distance from the current point of the method to the set of solutions with the increase in the number of iterations. The results of numerical experiments for the phase reconstruction problem (which is weakly convex and has a sharp minimum) are presented, demonstrating the effectiveness of the proposed approach to estimating the rate of convergence compared to the known one. Next, we propose a variation of the subgradient method with switching over productive and non-productive steps for weakly convex problems with inequality constraints and obtain the corresponding analog of the result on convergence with the rate of geometric progression. For the subgradient method with the corresponding variation of the Polyak stepsize on the class of relatively Lipschitz and relatively weakly convex functions with a relative analogue of a sharp minimum, it was obtained conditions that guarantee the convergence of such a subgradient method at the rate of a geometric progression. Finally, a theoretical result is obtained that describes the influence of the error of the information about the (sub)gradient available by the subgradient method and the objective function on the estimation of the quality of the obtained approximate solution. It is proved that for a sufficiently small error $\delta > 0$, one can guarantee that the accuracy of the solution is comparable to $\delta$.

This paper is devoted to some variants of improving the convergence rate guarantees of the gradient-type algorithms for relatively smooth and relatively Lipschitz-continuous problems in the case of additional information about some analogues of the strong convexity of the objective function. We consider two classes of problems, namely, convex problems with a relative functional growth condition, and problems (generally, non-convex) with an analogue of the Polyak – Lojasiewicz gradient dominance condition with respect to Bregman divergence. For the first type of problems, we propose two restart schemes for the gradient type methods and justify theoretical estimates of the convergence of two algorithms with adaptively chosen parameters corresponding to the relative smoothness or Lipschitz property of the objective function. The first of these algorithms is simpler in terms of the stopping criterion from the iteration, but for this algorithm, the near-optimal computational guarantees are justified only on the class of relatively Lipschitz-continuous problems. The restart procedure of another algorithm, in its turn, allowed us to obtain more universal theoretical results. We proved a near-optimal estimate of the complexity on the class of convex relatively Lipschitz continuous problems with a functional growth condition. We also obtained linear convergence rate guarantees on the class of relatively smooth problems with a functional growth condition. For a class of problems with an analogue of the gradient dominance condition with respect to the Bregman divergence, estimates of the quality of the output solution were obtained using adaptively selected parameters. We also present the results of some computational experiments illustrating the performance of the methods for the second approach at the conclusion of the paper. As examples, we considered a linear inverse Poisson problem (minimizing the Kullback – Leibler divergence), its regularized version which allows guaranteeing a relative strong convexity of the objective function, as well as an example of a relatively smooth and relatively strongly convex problem. In particular, calculations show that a relatively strongly convex function may not satisfy the relative variant of the gradient dominance condition.

The article is devoted to first-order adaptive methods for optimization problems with relatively strongly convex functionals. The concept of relatively strong convexity significantly extends the classical concept of convexity by replacing the Euclidean norm in the definition by the distance in a more general sense (more precisely, by Bregman’s divergence). An important feature of the considered classes of problems is the reduced requirements concerting the level of smoothness of objective functionals. More precisely, we consider relatively smooth and relatively Lipschitz-continuous objective functionals, which allows us to apply the proposed techniques for solving many applied problems, such as the intersection of the ellipsoids problem (IEP), the Support Vector Machine (SVM) for a binary classification problem, etc. If the objective functional is convex, the condition of relatively strong convexity can be satisfied using the problem regularization. In this work, we propose adaptive gradient-type methods for optimization problems with relatively strongly convex and relatively Lipschitzcontinuous functionals for the first time. Further, we propose universal methods for relatively strongly convex optimization problems. This technique is based on introducing an artificial inaccuracy into the optimization model, so the proposed methods can be applied both to the case of relatively smooth and relatively Lipschitz-continuous functionals. Additionally, we demonstrate the optimality of the proposed universal gradient-type methods up to the multiplication by a constant for both classes of relatively strongly convex problems. Also, we show how to apply the technique of restarts of the mirror descent algorithm to solve relatively Lipschitz-continuous optimization problems. Moreover, we prove the optimal estimate of the rate of convergence of such a technique. Also, we present the results of numerical experiments to compare the performance of the proposed methods.

Distributed saddle point problems (SPPs) have numerous applications in optimization, matrix games and machine learning. For example, the training of generated adversarial networks is represented as a min-max optimization problem, and training regularized linear models can be reformulated as an SPP as well. This paper studies distributed nonsmooth SPPs with Lipschitz-continuous objective functions. The objective function is represented as a sum of several components that are distributed between groups of computational nodes. The nodes, or agents, exchange information through some communication network that may be centralized or decentralized. A centralized network has a universal information aggregator (a server, or master node) that directly communicates to each of the agents and therefore can coordinate the optimization process. In a decentralized network, all the nodes are equal, the server node is not present, and each agent only communicates to its immediate neighbors.

We assume that each of the nodes locally holds its objective and can compute its value at given points, i. e. has access to zero-order oracle. Zero-order information is used when the gradient of the function is costly, not possible to compute or when the function is not differentiable. For example, in reinforcement learning one needs to generate a trajectory to evaluate the current policy. This policy evaluation process can be interpreted as the computation of the function value. We propose an approach that uses a smoothing technique, i. e., applies a first-order method to the smoothed version of the initial function. It can be shown that the stochastic gradient of the smoothed function can be viewed as a random two-point gradient approximation of the initial function. Smoothing approaches have been studied for distributed zero-order minimization, and our paper generalizes the smoothing technique on SPPs.

Non-smooth optimization often arises in many applied problems. The issues of developing efficient computational procedures for such problems in high-dimensional spaces are very topical. First-order methods (subgradient methods) are well applicable here, but in fairly general situations they lead to low speed guarantees for large-scale problems. One of the approaches to this type of problem can be to identify a subclass of non-smooth problems that allow relatively optimistic results on the rate of convergence. For example, one of the options for additional assumptions can be the condition of a sharp minimum, proposed in the late 1960s by B. T. Polyak. In the case of the availability of information about the minimal value of the function for Lipschitz-continuous problems with a sharp minimum, it turned out to be possible to propose a subgradient method with a Polyak step-size, which guarantees a linear rate of convergence in the argument. This approach made it possible to cover a number of important applied problems (for example, the problem of projecting onto a convex compact set). However, both the condition of the availability of the minimal value of the function and the condition of a sharp minimum itself look rather restrictive. In this regard, in this paper, we propose a generalized condition for a sharp minimum, somewhat similar to the inexact oracle proposed recently by Devolder – Glineur – Nesterov. The proposed approach makes it possible to extend the class of applicability of subgradient methods with the Polyak step-size, to the situation of inexact information about the value of the minimum, as well as the unknown Lipschitz constant of the objective function. Moreover, the use of local analogs of the global characteristics of the objective function makes it possible to apply the results of this type to wider classes of problems. We show the possibility of applying the proposed approach to strongly convex nonsmooth problems, also, we make an experimental comparison with the known optimal subgradient method for such a class of problems. Moreover, there were obtained some results connected to the applicability of the proposed technique to some types of problems with convexity relaxations: the recently proposed notion of weak $\beta$-quasi-convexity and ordinary quasiconvexity. Also in the paper, we study a generalization of the described technique to the situation with the assumption that the $\delta$-subgradient of the objective function is available instead of the usual subgradient. For one of the considered methods, conditions are found under which, in practice, it is possible to escape the projection of the considered iterative sequence onto the feasible set of the problem.

In the first part of the paper we present convergence analysis of AGMsDR method on a new class of functions — in general non-convex with $M$-Lipschitz-continuous gradients that satisfy Polyak – Lojasiewicz condition. Method does not need the value of $\mu^{PL}>0$ in the condition and converges linearly with a scale factor $\left(1 - \frac{\mu^{PL}}{M}\right)$. It was previously proved that method converges as $O\left(\frac1{k^2}\right)$ if a function is convex and has $M$-Lipschitz-continuous gradient and converges linearly with a~scale factor $\left(1 - \sqrt{\frac{\mu^{SC}}{M}}\right)$ if the value of strong convexity parameter $\mu^{SC}>0$ is known. The novelty is that one can save linear convergence if $\frac{\mu^{PL}}{\mu^{SC}}$ is not known, but without square root in the scale factor.

The second part presents modification of AGMsDR method for solving problems that allow alternating minimization (Alternating AGMsDR). The similar results are proved.

As the result, we present adaptive accelerated methods that converge as $O\left(\min\left\lbrace\frac{M}{k^2},\,\left(1-{\frac{\mu^{PL}}{M}}\right)^{(k-1)}\right\rbrace\right)$ on a class of convex functions with $M$-Lipschitz-continuous gradient that satisfy Polyak – Lojasiewicz condition. Algorithms do not need values of $M$ and $\mu^{PL}$. If Polyak – Lojasiewicz condition does not hold, the convergence is $O\left(\frac1{k^2}\right)$, but no tuning needed.

We also consider the adaptive catalyst envelope of non-accelerated gradient methods. The envelope allows acceleration up to $O\left(\frac1{k^2}\right)$. We present numerical comparison of non-accelerated adaptive gradient descent which is accelerated using adaptive catalyst envelope with AGMsDR, Alternating AGMsDR, APDAGD (Adaptive Primal-Dual Accelerated Gradient Descent) and Sinkhorn's algorithm on the problem dual to the optimal transport problem.

Conducted experiments show faster convergence of alternating AGMsDR in comparison with described catalyst approach and AGMsDR, despite the same asymptotic rate $O\left(\frac1{k^2}\right)$. Such behavior can be explained by linear convergence of AGMsDR method and was tested on quadratic functions. Alternating AGMsDR demonstrated better performance in comparison with AGMsDR.