/pdf/convex-analysisnoer-san-nojin-zhan-nituite-5dl2rbmoyr.pdf

Convex Analysisの二,三の進展について

In this paper, we analyze the convergence of the alternating direction method of multipliers (ADMM) for minimizing a nonconvex and possibly nonsmooth objective function, $$\phi (x_0,\ldots ,x_p,y)$$
 , subject to coupled linear equality constraints. Our ADMM updates each of the primal variables $$x_0,\ldots ,x_p,y$$
 , followed by updating the dual variable. We separate the variable y from $$x_i$$
 ’s as it has a special role in our analysis. The developed convergence guarantee covers a variety of nonconvex functions such as piecewise linear functions, $$\ell _q$$
 quasi-norm, Schatten-q quasi-norm (
 $$0<q<1$$
 ), minimax concave penalty (MCP), and smoothly clipped absolute deviation penalty. It also allows nonconvex constraints such as compact manifolds (e.g., spherical, Stiefel, and Grassman manifolds) and linear complementarity constraints. Also, the $$x_0$$
 -block can be almost any lower semi-continuous function. By applying our analysis, we show, for the first time, that several ADMM algorithms applied to solve nonconvex models in statistical learning, optimization on manifold, and matrix decomposition are guaranteed to converge. Our results provide sufficient conditions for ADMM to converge on (convex or nonconvex) monotropic programs with three or more blocks, as they are special cases of our model. ADMM has been regarded as a variant to the augmented Lagrangian method (ALM). We present a simple example to illustrate how ADMM converges but ALM diverges with bounded penalty parameter $$\beta $$
 . Indicated by this example and other analysis in this paper, ADMM might be a better choice than ALM for some nonconvex nonsmooth problems, because ADMM is not only easier to implement, it is also more likely to converge for the concerned scenarios.

http://www.optimization-online.org/DB_FILE/2015/11/5222.pdf

Global Convergence of ADMM in Nonconvex Nonsmooth Optimization

Recently, there have been growing interests in solving consensus optimization problems in a multi-agent network. In this paper, we develop a decentralized algorithm for the consensus optimization problem $$\min\limits_{x\in\mathbb{R}^p}~\bar{f}(x)=\frac{1}{n}\sum\limits_{i=1}^n f_i(x),$$ which is defined over a connected network of $n$ agents, where each function $f_i$ is held privately by agent $i$ and encodes the agent's data and objective. All the agents shall collaboratively find the minimizer while each agent can only communicate with its neighbors. Such a computation scheme avoids a data fusion center or long-distance communication and offers better load balance to the network. 
This paper proposes a novel decentralized EXact firsT-ordeR Algorithm (abbreviated as EXTRA) to solve the consensus optimization problem. "exact" means that it can converge to the exact solution. EXTRA can use a fixed large step size, {which is independent of the network size}, and has synchronized iterations. The local variable of every agent $i$ converges uniformly and consensually to an exact minimizer of $\bar{f}$. In contrast, the well-known decentralized gradient descent (DGD) method must use diminishing step sizes in order to converge to an exact minimizer. EXTRA and DGD have the same choice of mixing matrices and similar per-iteration complexity. EXTRA, however, uses the gradients of last two iterates, unlike DGD which uses just that of last iterate. 
EXTRA has the best known convergence rates among the existing first-order decentralized algorithms. Specifically, if $f_i$'s are convex and have Lipschitz continuous gradients, EXTRA has an ergodic convergence rate $O(\frac{1}{k})$ in terms of the first-order optimality residual. If $\bar{f}$ is also restricted strongly convex, EXTRA converges to an optimal solution at a linear rate $O(C^{-k})$ for some constant $C>1$.

EXTRA: An Exact First-Order Algorithm for Decentralized Consensus Optimization

The formulation $$\begin{aligned} \min _{x,y} ~f(x)+g(y),\quad \text{ subject } \text{ to } Ax+By=b, \end{aligned}$$minx,yf(x)+g(y),subjecttoAx+By=b,where f and g are extended-value convex functions, arises in many application areas such as signal processing, imaging and image processing, statistics, and machine learning either naturally or after variable splitting. In many common problems, one of the two objective functions is strictly convex and has Lipschitz continuous gradient. On this kind of problem, a very effective approach is the alternating direction method of multipliers (ADM or ADMM), which solves a sequence of f/g-decoupled subproblems. However, its effectiveness has not been matched by a provably fast rate of convergence; only sublinear rates such as O(1 / k) and $$O(1/k^2)$$O(1/k2) were recently established in the literature, though the O(1 / k) rates do not require strong convexity. This paper shows that global linear convergence can be guaranteed under the assumptions of strong convexity and Lipschitz gradient on one of the two functions, along with certain rank assumptions on A and B. The result applies to various generalizations of ADM that allow the subproblems to be solved faster and less exactly in certain manners. The derived rate of convergence also provides some theoretical guidance for optimizing the ADM parameters. In addition, this paper makes meaningful extensions to the existing global convergence theory of ADM generalizations.

/pdf/on-the-global-and-linear-convergence-of-the-generalized-4cx8t308zz.pdf

On the Global and Linear Convergence of the Generalized Alternating Direction Method of Multipliers

The alternating direction method of multipliers (ADMM) is widely used to solve large-scale linearly constrained optimization problems, convex or nonconvex, in many engineering fields. However there is a general lack of theoretical understanding of the algorithm when the objective function is nonconvex. In this paper we analyze the convergence of the ADMM for solving certain nonconvex consensus and sharing problems. We show that the classical ADMM converges to the set of stationary solutions, provided that the penalty parameter in the augmented Lagrangian is chosen to be sufficiently large. For the sharing problems, we show that the ADMM is convergent regardless of the number of variable blocks. Our analysis does not impose any assumptions on the iterates generated by the algorithm and is broadly applicable to many ADMM variants involving proximal update rules and various flexible block selection rules.

Convergence Analysis of Alternating Direction Method of Multipliers for a Family of Nonconvex Problems

Operator-splitting schemes are iterative algorithms for solving many types of numerical problems. A lot is known about these methods: they converge, and in many cases we know how quickly they converge. But when they are applied to optimization problems, there is a gap in our understanding: The theoretical speed of operator-splitting schemes is nearly always measured in the ergodic sense, but ergodic operator-splitting schemes are rarely used in practice. In this chapter, we tackle the discrepancy between theory and practice and uncover fundamental limits of a class of operator-splitting schemes. Our surprising conclusion is that the relaxed Peaceman-Rachford splitting algorithm, a version of the Alternating Direction Method of Multipliers (ADMM), is nearly as fast as the proximal point algorithm in the ergodic sense and nearly as slow as the subgradient method in the nonergodic sense. A large class of operator-splitting schemes extend from the relaxed Peaceman-Rachford splitting algorithm. Our results show that this class of operator-splitting schemes is also nearly as slow as the subgradient method. The tools we create in this chapter can also be used to prove nonergodic convergence rates of more general splitting schemes, so they are interesting in their own right.

http://www.optimization-online.org/DB_FILE/2014/06/4394.pdf

Convergence Rate Analysis of Several Splitting Schemes

Operator-splitting methods convert optimization and inclusion problems into fixed-point equations; when applied to convex optimization and monotone inclusion problems, the equations given by operator-splitting methods are often easy to solve by standard techniques. The hard part of this conversion, then, is to design nicely behaved fixed-point equations. In this paper, we design a new, and thus far, the only nicely behaved fixed-point equation for solving monotone inclusions with three operators; the equation employs resolvent and forward operators, one at a time, in succession. We show that our new equation extends the Douglas-Rachford and forward-backward equations; we prove that standard methods for solving the equation converge; and we give two accelerated methods for solving the equation.

A Three-Operator Splitting Scheme and its Optimization Applications

We consider a family of algorithms that successively sample and minimize simple stochastic models of the objective function. We show that under reasonable conditions on approximation quality and re...

Stochastic model-based minimization of weakly convex functions

This work considers the question: what convergence guarantees does the stochastic subgradient method have in the absence of smoothness and convexity? We prove that the stochastic subgradient method, on any semialgebraic locally Lipschitz function, produces limit points that are all first-order stationary. More generally, our result applies to any function with a Whitney stratifiable graph. In particular, this work endows the stochastic subgradient method, and its proximal extension, with rigorous convergence guarantees for a wide class of problems arising in data science---including all popular deep learning architectures.

Stochastic subgradient method converges on tame functions

This work considers the question: what convergence guarantees does the stochastic subgradient method have in the absence of smoothness and convexity? We prove that the stochastic subgradient method, on any semialgebraic locally Lipschitz function, produces limit points that are all first-order stationary. More generally, our result applies to any function with a Whitney stratifiable graph. In particular, this work endows the stochastic subgradient method, and its proximal extension, with rigorous convergence guarantees for a wide class of problems arising in data science—including all popular deep learning architectures.

Damek Davis

Papers

Convergence Rate Analysis of Several Splitting Schemes

A Three-Operator Splitting Scheme and its Optimization Applications

Stochastic model-based minimization of weakly convex functions

Stochastic subgradient method converges on tame functions

Stochastic Subgradient Method Converges on Tame Functions