Step-size Sequence Research Articles

Stochastic traffic assignment of mixed electric vehicle and gasoline vehicle flow with path distance constraints

Abstract: This paper addresses a general stochastic user equilibrium (SUE) traffic assignment problem (TAP) for transport networks with electric vehicles (EV), where EV paths are restricted by the EV driving range limits. A minimization model for path-constrained SUE is first proposed as an extension of path-constrained deterministic user equilibrium (DUE) TAP, which also extends the existing general SUE models with link-based constraints to path-based constraints. The resulting SUE model and solution algorithm can be used for other conditions with similar path-based constraints. The equilibrium conditions reveal that any path cost in the network is the sum of corresponding link costs and a path specific out-of-range penalty term, while path out-of-range term should equal to zero to ensure feasible flows. We develop a modified method of successive averages (MSA) with a predetermined step size sequence where both multinomial logit and multinomial probit based loading procedure are applied to solve the TAP. The suggested methods incorporate K-shortest paths algorithm to generate the path set on a need basis. Finally, two numerical examples are presented to verify the proposed model and solution algorithms.

Asymptotic behavior of truncated stochastic approximation procedures

We study asymptotic behavior of stochastic approximation procedures with three main characteristics: truncations with random moving bounds, a matrix-valued random step-size sequence, and a dynamically changing random regression function. In particular, we show that under quitemild conditions, stochastic approximation procedures are asymptotically linear in the statistical sense, that is, they can be represented as weighted sums of random variables. Therefore, a suitable formof the central limit theoremcan be applied to derive asymptotic distribution of the corresponding processes. The theory is illustrated by various examples and special cases.

Randomized Block Frank–Wolfe for Convergent Large-Scale Learning

Owing to their low-complexity iterations, Frank-Wolfe (FW) solvers are well suited for various large-scale learning tasks. When block-separable constraints are present, randomized block FW (RB-FW) has been shown to further reduce complexity by updating only a fraction of coordinate blocks per iteration. To circumvent the limitations of existing methods, the present work develops step sizes for RB-FW that enable a flexible selection of the number of blocks to update per iteration while ensuring convergence and feasibility of the iterates. To this end, convergence rates of RB-FW are established through computational bounds on a primal sub-optimality measure and on the duality gap. The novel bounds extend the existing convergence analysis, which only applies to a step-size sequence that does not generally lead to feasible iterates. Furthermore, two classes of step-size sequences that guarantee feasibility of the iterates are also proposed to enhance flexibility in choosing decay rates. The novel convergence results are markedly broadened to encompass also nonconvex objectives, and further assert that RB-FW with exact line-search reaches a stationary point at rate $\mathcal{O}(1/\sqrt{t})$. Performance of RB-FW with different step sizes and number of blocks is demonstrated in two applications, namely charging of electrical vehicles and structural support vector machines. Extensive simulated tests demonstrate the performance improvement of RB-FW relative to existing randomized single-block FW methods.

Rate of convergence of truncated stochastic approximation procedures with moving bounds

The paper is concerned with stochastic approximation procedures having three main characteristics: truncations with random moving bounds, a matrix valued random step-size sequence, and a dynamically changing random regression function. We study convergence and rate of convergence. Main results are supplemented with corollaries to establish various sets of sufficient conditions, with the main emphases on the parametric statistical estimation. The theory is illustrated by examples and special cases.

Adaptive burnup stepsize selection using control theory for 2-D lattice depletion simulations

A control theory approach is adopted to determine the temporal discretization during two-dimensional lattice physics depletion simulations. Two primary applications of automated and adaptive stepsize control are identified: (i) the presence of strong absorbers such as gadolinium, where the accurate burnout of the isotopes requires a depletion stepsize smaller than typically required, and (ii) high fidelity multiphysics simulations, e.g. loosely coupled physics, where the coupled physics are nonlinear in time and stepsize changes may be necessary to obtain an accurate coupled solution. A conventional predictor–corrector method is used to address the nonlinearity of the nuclide transmutation and neutron flux. An adaptive stepsize method is developed based on monitoring the one-group scalar neutron flux at both the predictor and corrector steps to approximate the convergence residual of the nonlinear solution. A user-specified tolerance on the L2 relative error norm of the scalar neutron flux is utilized by the stepsize controller. Controllers that include integral, proportional, and/or derivative components are investigated and parameterized using Latin hypercube sampling of the controller input parameters. Three distinct fuel loadings of pressurized water reactor 17 × 17 fuel pin assemblies are considered, including no burnable absorbers, Integral Fuel Burnable Absorber, and gadolinium fuel pins. The required depletion stepsizes, as predicted throughout the cycle by the controller, are compared with a very small stepsize (0.01 MW d/kgHM) reference solution and a solution obtained by a typical rule of thumb depletion stepsize sequence.

Consistency and fluctuations for stochastic gradient Langevin dynamics

Applying standard Markov chain Monte Carlo (MCMC) algorithms to large data sets is computationally expensive. Both the calculation of the acceptance probability and the creation of informed proposals usually require an iteration through the whole data set. The recently proposed stochastic gradient Langevin dynamics (SGLD) method circumvents this problem by generating proposals which are only based on a subset of the data, by skipping the accept-reject step and by using decreasing step-sizes sequence (δm)m≥0. We provide in this article a rigorous mathematical framework for analysing this algorithm. We prove that, under verifiable assumptions, the algorithm is consistent, satisfies a central limit theorem (CLT) and its asymptotic bias-variance decomposition can be characterized by an explicit functional of the step-sizes sequence (δm)m≥0. We leverage this analysis to give practical recommendations for the notoriously difficult tuning of this algorithm: it is asymptotically optimal to use a step-size sequence of the type δm = m-1/3, leading to an algorithm whose mean squared error (MSE) decreases at rate O(m-1/3).

Parallel computing subgradient method for nonsmooth convex optimization over the intersection of fixed point sets of nonexpansive mappings

Nonsmooth convex optimization problems are solved over fixed point sets of nonexpansive mappings by using a distributed optimization technique. This is done for a networked system with an operator, who manages the system, and a finite number of users, by solving the problem of minimizing the sum of the operator’s and users’ nondifferentiable, convex objective functions over the intersection of the operator’s and users’ convex constraint sets in a real Hilbert space. We assume that each of their constraint sets can be expressed as the fixed point set of an implementable nonexpansive mapping. This setting allows us to discuss nonsmooth convex optimization problems in which the metric projection onto the constraint set cannot be calculated explicitly. We propose a parallel subgradient algorithm for solving the problem by using the operator’s attribution such that it can communicate with all users. The proposed algorithm does not use any proximity operators, in contrast to conventional parallel algorithms for nonsmooth convex optimization. We first study its convergence property for a constant step-size rule. The analysis indicates that the proposed algorithm with a small constant step size approximates a solution to the problem. We next consider the case of a diminishing step-size sequence and prove that there exists a subsequence of the sequence generated by the algorithm which weakly converges to a solution to the problem. We also give numerical examples to support the convergence analyses.

Extremum estimation and numerical derivatives

Finite-difference approximations are widely used in empirical work to evaluate derivatives of estimated functions. For instance, many standard optimization routines rely on finite-difference formulas for gradient calculations and estimating standard errors. However, the effect of such approximations on the statistical properties of the resulting estimators has only been studied in a few special cases. This paper investigates the impact of commonly used finite-difference methods on the large sample properties of the resulting estimators. We find that first, one needs to adjust the step size as a function of the sample size. Second, higher-order finite difference formulas reduce the asymptotic bias analogous to higher order kernels. Third, we provide weak sufficient conditions for uniform consistency of the finite-difference approximations for gradients and directional derivatives. Fourth, we analyze numerical gradient-based extremum estimators and find that the asymptotic distribution of the resulting estimators may depend on the sequence of step sizes. We state conditions under which the numerical derivative based extremum estimator is consistent and asymptotically normal. Fifth, we generalize our results to semiparametric estimation problems. Finally, we demonstrate that our results apply to a range of nonstandard estimation procedures.

PARALLEL OPTIMIZATION ALGORITHM FOR SMOOTH CONVEX OPTIMIZATION OVER FIXED POINT SETS OF QUASI-NONEXPANSIVE MAPPINGS

Smooth convex optimization problems are solved over fixed point sets of quasi-nonexpansive mappings by using a distributed optimization technique. This is done for a networked system with an operator, who manages the system, and a finite number of users, by solving the problem of minimizing the sum of the operator's and users' differentiable, convex objective functions over the intersection of the operator's and users' fixed point sets of quasi-nonexpansive mappings. Under the assumption that the operator can communicate with all users, a parallel optimization algorithm can be devised that enables the operator to find a solution to the problem without using all user objective functions and quasi-nonexpansive mappings. This algorithm does not use proximity operators, in contrast to conventional parallel proximal algorithms. Moreover, it can optimize over fixed point sets of quasi-nonexpansive mappings, in contrast to conventional fixed point algorithms. Investigation of the algorithm's convergence properties for a constant step-size rule reveals that, with a small constant step size, it approximates the solution to the problem. Consideration of the case in which the step-size sequence is diminishing demonstrates that the algorithm converges to the problem solution. Application of the algorithm to network bandwidth allocation based on an operational policy is shown to make the network more stable and reliable.

New analysis and results for the Frank–Wolfe method

We present new results for the Frank---Wolfe method (also known as the conditional gradient method). We derive computational guarantees for arbitrary step-size sequences, which are then applied to various step-size rules, including simple averaging and constant step-sizes. We also develop step-size rules and computational guarantees that depend naturally on the warm-start quality of the initial (and subsequent) iterates. Our results include computational guarantees for both duality/bound gaps and the so-called FW gaps. Lastly, we present complexity bounds in the presence of approximate computation of gradients and/or linear optimization subproblem solutions.

Online Learning as Stochastic Approximation of Regularization Paths: Optimality and Almost-Sure Convergence

In this paper, an online learning algorithm is proposed as sequential stochastic approximation of a regularization path converging to the regression function in reproducing kernel Hilbert spaces (RKHSs). We show that it is possible to produce the best known strong (RKHS norm) convergence rate of batch learning, through a careful choice of the gain or step size sequences, depending on regularity assumptions on the regression function. The corresponding weak (mean square distance) convergence rate is optimal in the sense that it reaches the minimax and individual lower rates in the literature. In both cases we deduce almost sure convergence, using Bernstein-type inequalities for martingales in Hilbert spaces. To achieve this we develop a bias-variance decomposition similar to the batch learning setting; the bias consists in the approximation and drift errors along the regularization path, which display the same rates of convergence, and the variance arises from the sample error analysed as a (reverse) martingale difference sequence. The rates above are obtained by an optimal trade-off between the bias and the variance.

Multi-level stochastic approximation algorithms

This paper studies multi-level stochastic approximation algorithms. Our aim is to extend the scope of the multilevel Monte Carlo method recently introduced by Giles (Giles 2008) to the framework of stochastic optimization by means of stochastic approximation algorithm. We first introduce and study a two-level method, also referred as statistical Romberg stochastic approximation algorithm. Then, its extension to multi-level is proposed. We prove a central limit theorem for both methods and describe the possible optimal choices of step size sequence. Numerical results confirm the theoretical analysis and show a significant reduction in the initial computational cost.

Markovian stochastic approximation with expanding projections

Stochastic approximation is a framework unifying many random iterative algorithms occurring in a diverse range of applications. The stability of the process is often difficult to verify in practical applications and the process may even be unstable without additional stabilisation techniques. We study a stochastic approximation procedure with expanding projections similar to Andradóttir [Oper. Res. 43 (1995) 1037–1048]. We focus on Markovian noise and show the stability and convergence under general conditions. Our framework also incorporates the possibility to use a random step size sequence, which allows us to consider settings with a non-smooth family of Markov kernels. We apply the theory to stochastic approximation expectation maximisation with particle independent Metropolis–Hastings sampling.

Truncated stochastic approximation with moving bounds: convergence

In this paper we consider a wide class of truncated stochastic approximation procedures. These procedures have three main characteristics: truncations with random moving bounds, a matrix valued random step-size sequence, and a dynamically changing random regression function. We establish convergence and consider several examples to illustrate the results.

Step-size sequence design for finite-time average consensus in secure wireless sensor networks

This paper concerns the study of average consensus in wireless sensor networks with aim of providing a way to reach consensus in a finite number of steps. In particular, we investigate the design of consensus protocols when, for security reasons for instance, the underlying graph is constrained to be strongly regular or distance regular. The proposed design method is based on parameters of the intersection array characterizing the underlying graph. With this protocol, at execution time, average consensus is achieved in a number of steps equal to the diameter of the graph, i.e. the smallest possible number of steps to achieve consensus.

Approximate solutions to variational inequality over the fixed point set of a strongly nonexpansive mapping

Abstract Variational inequality problems over fixed point sets of nonexpansive mappings include many practical problems in engineering and applied mathematics, and a number of iterative methods have been presented to solve them. In this paper, we discuss a variational inequality problem for a monotone, hemicontinuous operator over the fixed point set of a strongly nonexpansive mapping on a real Hilbert space. We then present an iterative algorithm, which uses the strongly nonexpansive mapping at each iteration, for solving the problem. We show that the algorithm potentially converges in the fixed point set faster than algorithms using firmly nonexpansive mappings. We also prove that, under certain assumptions, the algorithm with slowly diminishing step-size sequences converges to a solution to the problem in the sense of the weak topology of a Hilbert space. Numerical results demonstrate that the algorithm converges to a solution to a concrete variational inequality problem faster than the previous algorithm. MSC:47H06, 47J20, 47J25.

Acceleration method for convex optimization over the fixed point set of a nonexpansive mapping

The existing algorithms for solving the convex minimization problem over the fixed point set of a nonexpansive mapping on a Hilbert space are based on algorithmic methods, such as the steepest descent method and conjugate gradient methods, for finding a minimizer of the objective function over the whole space, and attach importance to minimizing the objective function as quickly as possible. Meanwhile, it is of practical importance to devise algorithms which converge in the fixed point set quickly because the fixed point set is the set with the constraint conditions that must be satisfied in the problem. This paper proposes an algorithm which not only minimizes the objective function quickly but also converges in the fixed point set much faster than the existing algorithms and proves that the algorithm with diminishing step-size sequences strongly converges to the solution to the convex minimization problem. We also analyze the proposed algorithm with each of the Fletcher---Reeves, Polak---Ribiere---Polyak, Hestenes---Stiefel, and Dai---Yuan formulas used in the conventional conjugate gradient methods, and show that there is an inconvenient possibility that their algorithms may not converge to the solution to the convex minimization problem. We numerically compare the proposed algorithm with the existing algorithms and show its effectiveness and fast convergence.

An infeasible-point subgradient method using adaptive approximate projections

We propose a new subgradient method for the minimization of nonsmooth convex functions over a convex set. To speed up computations we use adaptive approximate projections only requiring to move within a certain distance of the exact projections (which decreases in the course of the algorithm). In particular, the iterates in our method can be infeasible throughout the whole procedure. Nevertheless, we provide conditions which ensure convergence to an optimal feasible point under suitable assumptions. One convergence result deals with step size sequences that are fixed a priori. Two other results handle dynamic Polyak-type step sizes depending on a lower or upper estimate of the optimal objective function value, respectively. Additionally, we briefly sketch two applications: Optimization with convex chance constraints, and finding the minimum ℓ 1-norm solution to an underdetermined linear system, an important problem in Compressed Sensing.

Asymptotic optimality for consensus-type stochastic approximation algorithms using iterate averaging

This paper introduces a post-iteration averaging algorithm to achieve asymptotic optimality in convergence rates of stochastic approximation algorithms for consensus control with structural constraints. The algorithm involves two stages. The first stage is a coarse approximation obtained using a sequence of large stepsizes. Then, the second stage provides a refinement by averaging the iterates from the first stage. We show that the new algorithm is asymptotically efficient and gives the optimal convergence rates in the sense of the best scaling factor and ‘smallest’ possible asymptotic variance.

Fixed Point Optimization Algorithms for Distributed Optimization in Networked Systems

This paper considers a networked system with a finite number of users and deals with the problem of minimizing the sum of all users' objective functions over the intersection of all users' constraint sets, onto which the projection cannot be easily implemented. The main objective of this paper is to devise distributed optimization algorithms, which enable each user to find the solution of the problem without using other users' objective functions and constraint sets. To reach this goal, we first introduce easily implementable nonexpansive mappings of which the intersection of the fixed point sets is equal to the constraint set in the problem. We formulate the problem as a convex minimization problem over the intersection of the fixed point sets of the nonexpansive mappings. We then present an iterative algorithm, based on the conventional incremental subgradient methods which use the projection, for solving the problem. The algorithm can be implemented by using nonexpansive mappings other than the projection. We prove that the algorithm with slowly diminishing step-size sequences converges to a solution of the problem in the sense of weak topology of a Hilbert space. We also present a broadcast type of distributed optimization algorithm that weakly converges to a solution of the problem. Numerical examples for the bandwidth allocation demonstrate the convergence of these algorithms.