DOCSLIB.ORG

Simulated Annealing a Basic Introduction

Total Page：16

File Type：pdf, Size：1020Kb

Download full-text PDF Read full-text
Simulated Annealing a Basic Introduction Simulated Annealing A Basic Introduction Prof. Olivier de Weck Massachusetts Institute of Technology Dr. Cyrus Jilla Outline • Heuristics • Background in Statistical Mechanics – Atom Configuration Problem – Metropolis Algorithm • Simulated Annealing Algorithm • Sample Problems and Applications • Summary Learning Objectives • Review background in Statistical Mechanics: configuration, ensemble, entropy, heat capacity • Understand the basic assumptions and steps in Simulated Annealing • Be able to transform design problems into a combinatorial optimization problem suitable to SA • Understand strengths and weaknesses of SA Heuristics What is a Heuristic? • A Heuristic is simply a rule of thumb that hopefully will find a good answer. • Why use a Heuristic? – Heuristics are typically used to solve complex (large, nonlinear, nonconvex (ie. contain many local minima)) multivariate combinatorial optimization problems that are difficult to solve to optimality. • Unlike gradient-based methods in a convex design space, heuristics are NOT guaranteed to find the true global optimal solution in a single objective problem, but should find many good solutions (the mathematician's answer vs. the engineer’s answer) • Heuristics are good at dealing with local optima without getting stuck in them while searching for the global optimum. Types of Heuristics • Heuristics Often Incorporate Randomization • 2 Special Cases of Heuristics – Construction Methods • Must first find a feasible solution and then improve it. – Improvement Methods • Start with a feasible solution and just try to improve it. • 3 Most Common Heuristic Techniques – Genetic Algorithms – Simulated Annealing – Tabu Search – New Methods: Particle Swarm Optimization, etc… Origin of Simulated Annealing (SA) • Definition: A heuristic technique that mathematically mirrors the cooling of a set of atoms to a state of minimum energy. • Origin: Applying the field of Statistical Mechanics to the field of Combinatorial Optimization (1983) • Draws an analogy between the cooling of a material (search for minimum energy state) and the solving of an optimization problem. • Original Paper Introducing the Concept – Kirkpatrick, S., Gelatt, C.D., and Vecchi, M.P., “Optimization by Simulated Annealing,” Science, Volume 220, Number 4598, 13 May 1983, pp. 671­ 680. Statistical Mechanics The Analogy • Statistical Mechanics: The behavior of systems with many degrees of freedom in thermal equilibrium at a finite temperature. • Combinatorial Optimization: Finding the minimum of a given function depending on many variables. • Analogy: If a liquid material cools and anneals too quickly, then the material will solidify into a sub-optimal configuration. If the liquid material cools slowly, the crystals within the material will solidify optimally into a state of minimum energy (i.e. ground state). – This ground state corresponds to the minimum of the cost function in an optimization problem. Sample Atom Configuration Original Configuration Perturbed Configuration Atom Configuration - Sample Problem Atom Configuration - Sample Problem 7 7 6 6 5 5 4 3 4 3 4 3 4 2 2 2 3 2 1 1 1 1 0 0 -1 -1 -1 0 1 2 3 4 5 6 7 -1 0 1 2 3 4 5 6 7 E=133.67 E=109.04 Perturbing = move a random atom Energy of original (configuration) to a new random (unoccupied) slot Configurations • Mathematically describe a configuration – Specify coordinates of each “atom” Ø1ø Ø 3 øØ 4 ø Øø 5 {ri } with i=1,..,4 r1 = Œœ, r2 = Œ œ, r3 = Œ œ, r4 = Œœ º1ß º2 ß º4 ß ºß3 – Specify a slot for each atom {ri} with i=1,..,4 R = [1 12 19 23] Energy of a state • Each state (configuration) has an energy level associated with it H( qq,�,t) = � q�iip- L( qqt ,,� ) i Hamiltonian H =+=T VEtot Energy of configuration = Objective function value of design kinetic potential Energy Energy sample problem • Define energy function for “atom” sample problem 1 N 2 2 E = mgy + x - x + y - y 2 i i �(( i j ) ( i j ) ) potential� ji„ energy ������������� kinetic energy N Absolute and relative position of E( R) = � Eri ( i ) i=1 each atom contributes to Energy Compute Energy of Config. A • Energy of initial configuration E1 = 1� 10� 1+ 5 + 18+ 20= 20.95 E2 = 1� 10� 2+ 5 + 5 + 5 = 26.71 E3 = 1� 10� 4+ 18+ 5 + 2 = 47.89 E4 = 1� 10� 3+ 20+ 5 + 2 = 38.12 Total Energy Configuration A: E({rA})= 133.67 Boltzmann Probability P! Number of configurations NR = ( PN- )! P=# of slots=25 N =6,375,600 N=# of atoms =4 R What is the likelihood that a particular configuration will exist in a large ensemble of configurations? Ø-Er({ }) ø Boltzmann probability Pr({ }) = exp Œ œ depends on energy º kTB ß and temperature Boltzmann Collapse at low T T=100 T=10 T=1 T=100 - E =100.1184 T=10 - E =58.9245 avg avg T=1 - E avg =38.1017 14000 14000 12000 12000 12000 10000 10000 10000 8000 8000 8000 6000 Occurences 6000 Occurences 6000 Occurences 4000 4000 4000 2000 2000 2000 0 0 20 40 60 80 100 120 140 160 180 200 0 0 20 40 60 80 100 120 140 160 180 200 0 Energy 0 20 40 60 80 100 120 140 160 180 200 Energy Energy T high T low Boltzmann Distribution collapses to the lowest energy state(s) in the limit of low temperature Basis of search by Simulated Annealing Partition Function Z • Ensemble of configurations can be described statistically • Partition function, Z, generates the ensemble average � -E � NR � -E � Z = Tr exp� i � = � exp � i � Ł kB Tł i=1 Ł kTB ł Initially (at T>>0) equal to the number of possible configurations Free Energy NR Average Energy of all Eav g = E()T= � ET i ( ) Configurations in an i=1 ensemble F( T) =- kB TZln= E ()T- TS NR � ETi ( ) � �exp � - � ETi ( ) i=1 Ł kTB ł -dZln E avg = ET() = = Z d (1 kTB ) Relates average Energy at T with Entropy S Specific Heat and Entropy • Specific Heat NR Ø 1 2 2 ø Œ � Ei () T- ET ( ) œ 2 2 dET() º N R i=1 ß E() T- ET ( ) CT( ) = = 2 = 2 dT kBT kTB • Entropy T 1 CT() dS(T)CT () S(T)= S ()T- dT 1 � = T T dTT Entropy is ~ equal to ln(# of unique configurations) Low Temperature Statistics Specific Heat C and Entropy S 5 10 10 T 8 Entropy S s, N 6 decreases 4 with lower T onfiguration f c 2 # o 0 # configs 10 0 Ht. Capacity 0 2 4 0 2 4 10 10 10 10 10 10 decreases Temperature Temperature 150 150 100 g 100 v a 50 y, E 50 y, F 0 0 Energ -50 -50 Free Energ erage -100 -100 Approaches Approaches Av -150 -150 E* from E* from 0 2 4 0 2 4 10 10 10 10 10 10 below Temperature Temperature above Minimum Energy Configurations • Sample Atom Placement Problem Uniqueness? Atom Configuration - Sample Problem Atom Configuration - Sample Problem 7 7 6 6 5 E*=60 5 E*=14.26 4 4 3 3 2 2 3 2 1 2 1 4 3 1 4 1 0 0 -1 -1 -1 0 1 2 3 4 5 6 7 -1 0 1 2 3 4 5 6 7 Strong gravity g=10 Low gravity g=0.1 Simulated Annealing Dilemma • Cannot compute energy of all configurations ! – Design space often too large – Computation time for a single function evaluation can be large • Use Metropolis Algorithm, at successively lower temperatures to find low energy states – Metropolis: Simulate behavior of a set of atoms in thermal equilibrium (1953) – Probability of a configuration existing at T � Boltzmann Probability P(r,T)=exp(-E(r)/T) The SA Algorithm • Terminology: – X (or R or G) = Design Vector (i.e. Design, Architecture, Configuration) – E = System Energy (i.e. Objective Function Value) – T = System Temperature – D = Difference in System Energy Between Two Design Vectors • The Simulated Annealing Algorithm 1) Choose a random Xi, select the initial system temperature, and specify the cooling (i.e. annealing) schedule 2) Evaluate E(Xi) using a simulation model 3) Perturb Xi to obtain a neighboring Design Vector (Xi+1) 4) Evaluate E(Xi+1) using a simulation model 5) If E(Xi+1)< E(Xi), Xi+1 is the new current solution 6) If E(Xi+1)> E(Xi), then accept Xi+1 as the new current solution with a (- D/T) probability e where D = E(Xi+1) - E(Xi). 7) Reduce the system temperature according to the cooling schedule. 8) Terminate the algorithm. SA BLOCK DIAGRAM Start To End Tmin � � � y Define Initial n T < Tmin ? Configuration Ro � Perturb Configuration Reduce Temperature Evaluate Energy E(Ro) � Ri -> Ri+1 � Tj+1 = Tj - �T �� �� n y � Evaluate Energy E(Ri+1) Reached Equilibrium at Tj ? � Compute Energy Difference Accept R as i+1 Keep Ri as Current �E = E(Ri+1) - E(Ri) � New Configuration Configuration � y y n �E < 0 ? exp -�E > � ? � � Metropolis n Step � � Create Random Number � in [0,1] Matlab Function: SA.m Best Configuration Initial Configuration SA.m xbest Ebest xo Simulated Annealing Option Flags Algorithm History options xhist evaluation perturbation function function file_eval.m file_perturb.m Exponential Cooling k • Typically (T1/To)~0.7-0.9 � T � T = 1 T Simulated Annealing Evolution k +1 � � k 10 C-specific heat Ł To ł 9 S-entropy T/10-temperature 8 7 Can also do linear 6 or step-wise cooling 5 4 3 … 2 But exponential cooling 1 often works best. 0 0 5 10 15 20 25 30 35 40 45 Temperature Step Key Ingredients for SA • A concise description of a configuration (architecture, design, topology) of the system (Design Vector). • A random generator of rearrangements of the elements in a configuration (Neighborhoods).
Load more

Recommended publications
  • Simulated Annealing Hyper-Heuristic for a Shelf Space Allocation on Symmetrical Planograms Problem
    S S symmetry Article Simulated Annealing Hyper-Heuristic for a Shelf Space Allocation on Symmetrical Planograms Problem Kateryna Czerniachowska * and Marcin Hernes Department of Process Management, Wroclaw University of Economics and Business, Komandorska 118/120, 53-345 Wroclaw, Poland; [email protected] * Correspondence: [email protected] Abstract: The allocation of products on shelves is an important issue from the point of view of effective decision making by retailers. In this paper, we investigate a practical shelf space allocation model which takes into account the number of facings, capping, and nesting of a product. We divide the shelf into the segments of variable size in which the products of the specific types could be placed. The interconnections between products are modelled with the help of categorizing the products into specific types as well as grouping some of them into clusters. This results in four groups of constraints—shelf constraints, shelf type constraints, product constraints, position allocation constraints—that are used in the model for aesthetic symmetry of a planogram. We propose a simulated annealing algorithm with improvement and reallocation procedures to solve the planogram profit maximization problem. Experiments are based on artificial data sets that have been generated according to real-world conditions. The efficiency of the designed algorithm has been estimated using the CPLEX solver. The computational tests demonstrate that the proposed algorithm gives valuable results in an acceptable time. Citation: Czerniachowska, K.; Hernes, M. Simulated Annealing Keywords: retailing decisions; shelf space allocation; simulated annealing; heuristics; merchandising Hyper-Heuristic for a Shelf Space Allocation on Symmetrical Planograms Problem.
    [Show full text]
  • A Simulated Annealing Based Inexact Oracle for Wasserstein Loss Minimization
    A Simulated Annealing Based Inexact Oracle for Wasserstein Loss Minimization Jianbo Ye 1 James Z. Wang 1 Jia Li 2 Abstract ryl(x; ·) is computed in O(m) time, m being the complex- ity of outcome variables x or y. This part of calculation is Learning under a Wasserstein loss, a.k.a. Wasser- often negligible compared with the calculation of full gra- stein loss minimization (WLM), is an emerging dient with respect to the model parameters. But this is no research topic for gaining insights from a large longer the case in learning problems based on Wasserstein set of structured objects. Despite being concep- distance due to the intrinsic complexity of the distance. tually simple, WLM problems are computation- We will call such problems Wasserstein loss minimization ally challenging because they involve minimiz- (WLM). Examples of WLMs include Wasserstein barycen- ing over functions of quantities (i.e. Wasserstein ters (Li & Wang, 2008; Agueh & Carlier, 2011; Cuturi & distances) that themselves require numerical al- Doucet, 2014; Benamou et al., 2015; Ye & Li, 2014; Ye gorithms to compute. In this paper, we intro- et al., 2017b), principal geodesics (Seguy & Cuturi, 2015), duce a stochastic approach based on simulated nonnegative matrix factorization (Rolet et al., 2016; San- annealing for solving WLMs. Particularly, we dler & Lindenbaum, 2009), barycentric coordinate (Bon- have developed a Gibbs sampler to approximate neel et al., 2016), and multi-label classification (Frogner effectively and efficiently the partial gradients of et al., 2015). a sequence of Wasserstein losses. Our new ap- proach has the advantages of numerical stability Wasserstein distance is defined as the cost of matching two and readiness for warm starts.
    [Show full text]
  • University of Waterloo Faculty of Engineering
    University of Waterloo Faculty of Engineering Global Optimization of General Failure Surfaces in Slope Analysis by Hybrid Simulated Annealing Rocscience, Inc. Toronto, Ontario Prepared by Su, Xiao ID 20251755 2B Chemical Engineering May 1st , 2009 4105 Credit Pointe Dr, Mississauga, Ontario L5M 3K7 May 1st, 2009 Dr Thomas Duever, Chair of Chemical Engineering University of Waterloo Waterloo, Ontario N2L 3G1 Dear Dr Duever, I am submitting this document, entitled “Global Optimization of General Failure Surfaces in Slope Analysis by Hybrid Simulated Annealing” as my 2B Work Report for Rocscience, Inc. My project investigated the efficiency of a hybrid simulated annealing (HSA) method for finding the minima (critical failure surfaces and factors of safety) of slope stability equations. The HSA combines a Very Fast Simulated Annealing (VFSA) and a Local Monte-Carlo (LMC) search, the latter being an optimization algorithm designed specifically for solving slope stability problems. This work is an extension as well as a generalization of the project from my summer work-term (2008), also undertaken at Rocscience. During both terms, I worked under the supervision of Dr. Brent Corkum, software development manager at Rocscience. I implemented the simulated annealing algorithm as a new optimization method in SLIDE 5.0, a slope stability analysis program developed by Rocscience. In addition to learning professional C++ and principles of object oriented programming, I had the opportunity to study the theoretical and practical aspects of non-convex optimization, especially the performance of stochastic methods when used on multi- modal functions. I also learned soil mechanics and slope stability principles. This report was written entirely by me and has not received any previous academic credit at Rocscience or the University of Waterloo.
    [Show full text]
  • Arxiv:2101.10154V1 [Cond-Mat.Dis-Nn] 25 Jan 2021
    Variational Neural Annealing 1, 2, 3, 1 1, 2 2, 3 1, 2 Mohamed Hibat-Allah, ∗ Estelle M. Inack, Roeland Wiersema, Roger G. Melko, and Juan Carrasquilla 1Vector Institute, MaRS Centre, Toronto, Ontario, M5G 1M1, Canada 2Department of Physics and Astronomy, University of Waterloo, Ontario, N2L 3G1, Canada 3Perimeter Institute for Theoretical Physics, Waterloo, ON N2L 2Y5, Canada (Dated: January 26, 2021) Many important challenges in science and technology can be cast as optimization problems. When viewed in a statistical physics framework, these can be tackled by simulated annealing, where a gradual cooling procedure helps search for groundstate solutions of a target Hamiltonian. While powerful, simulated annealing is known to have prohibitively slow sampling dynamics when the optimization landscape is rough or glassy. Here we show that by generalizing the target distribution with a parameterized model, an analogous annealing framework based on the variational principle can be used to search for groundstate solutions. Modern autoregressive models such as recurrent neural networks provide ideal parameterizations since they can be exactly sampled without slow dynamics even when the model encodes a rough landscape. We implement this procedure in the classical and quantum settings on several prototypical spin glass Hamiltonians, and find that it significantly outperforms traditional simulated annealing in the asymptotic limit, illustrating the potential power of this yet unexplored route to optimization. P I. INTRODUCTION <latexit sha1_base64="Wk+wWdaEIxpmR5GD8uDFICvy2Ig=">AAACNHicfVDLSgMxFL1TX7W+Rl26CRbBVZkRQZdFN4KbCvYBbRkyaaYNzWSGJGMpQz/KjR/iRgQXirj1G8y0s6iteCBwOOfcJPf4MWdKO86rVVhZXVvfKG6WtrZ3dvfs/YOGihJJaJ1EPJItHyvKmaB1zTSnrVhSHPqcNv3hdeY3H6hULBL3ehzTboj7ggWMYG0kz76teSnq9KKRwFJGI9RJ4kUyZ/6Xm5Q8u+xUnCnQMnFzUoYcNc9+NjeSJKRCE46VartOrLsplpoRTielTqJojMkQ92nbUIFDqrrpdOkJOjFKDwWRNEdoNFXnJ1IcKjUOfZMMsR6oRS8T//LaiQ4uuykTcaKpILOHgoQjHaGsQdRjkhLNx4ZgIpn5KyIDLDHRpuesBHdx5WXSOKu4TsW9Oy9Xr/I6inAEx3AKLlxAFW6gBnUg8Agv8A4f1pP1Zn1aX7NowcpnDuEXrO8fzwWstw==</latexit>
    [Show full text]
  • A Simulated Annealing Algorithm for Solving Two-Echelon Vehicle Routing Problem with Locker Facilities
    algorithms Article A Simulated Annealing Algorithm for Solving Two-Echelon Vehicle Routing Problem with Locker Facilities A. A. N. Perwira Redi 1 , Parida Jewpanya 2,*, Adji Candra Kurniawan 3, Satria Fadil Persada 4, Reny Nadlifatin 5 and Oki Anita Candra Dewi 6 1 Industrial Engineering Department, BINUS Graduate Program—Master of Industrial Engineering, Bina Nusantara University, Jakarta 11480, Indonesia; [email protected] 2 Department of Industrial Engineering, Rajamangala University of Technology Lanna Tak, Tak 63000, Thailand 3 Department of Logistics Engineering, Universitas Pertamina, Jakarta 12220, Indonesia; [email protected] 4 Department of Business Management, Institut Teknologi Sepuluh Nopember, Surabaya 60111, Indonesia; [email protected] 5 Department of Technology Management, Institut Teknologi Sepuluh Nopember, Surabaya 60111, Indonesia; [email protected] 6 Logistics Engineering Department, Universitas Internasional Semen Indonesia, Gresik 61122, Indonesia; [email protected] * Correspondence: [email protected] Received: 6 August 2020; Accepted: 1 September 2020; Published: 3 September 2020 Abstract: We consider the problem of utilizing the parcel locker network for the logistics solution in the metropolitan area. Two-echelon distribution systems are attractive from an economic standpoint, whereas the product from the depot can be distributed from or to intermediate facilities. In this case, the intermediate facilities are considered as locker facilities present in an accessible location in the vicinity of the final customers. In addition, the utilization of locker facilities can reduce the cost caused by the unattended deliveries. The problem is addressed as an optimization model that formulated into an integer linear programming model denoted as the two-echelon vehicle routing problem with locker facilities (2EVRP-LF).
    [Show full text]
  • Microwave Imaging for Conducting Scatterers by Hybrid Particle Swarm Optimization with Simulated Annealing
    2011 8th International Multi-Conference on Systems, Signals & Devices MICROWAVE IMAGING FOR CONDUCTING SCATTERERS BY HYBRID PARTICLE SWARM OPTIMIZATION WITH SIMULATED ANNEALING Bouzid Mhamdi, Khaled Grayaa and Taoufik Aguili Syscom Laboratory, Engineer School of Tunis - BP 37 Belvedere, Tunis - 1002, Tunisia [email protected] ABSTRACT The first is based on gradient searching schemes such as the Newton-Kantorovitch method [3], modified In this paper, a microwave imaging technique for gradient method [4] and Levenberg-Marguart algorithm reconstructing the shape of two-dimensional perfectly [5]. It is well-known that these deterministic techniques conducting scatterers by means of a stochastic used for fast reconstruction of microwave images tend to optimization approach is investigated. get trapped in a local extreme and suffer from a major Based on the boundary condition and the measured drawback, where the final image is highly dependent on scattered field derived by transverse magnetic the initial trial solution. To overcome this obstacle, a illuminations, a set of nonlinear integral equations is second class of population based stochastic methods such obtained and the imaging problem is reformulated in to as the genetic algorithm (GA) [1-2], simulated annealing an optimization problem. (SA) [6] and neural network (NN) [7] has become A hybrid approximation algorithm, called PSO-SA, is attractive alternatives to reconstruct microwave images. developed in this work to solve the scattering inverse These techniques consider the imaging problem as a problem. In the hybrid algorithm, particle swarm optimization (PSO) combines global search and local global optimization problem, and reconstruct the correct search for finding the optimal results assignment with image by searching for the optimal solution through the reasonable time and simulated annealing (SA) uses use of either rivalry or cooperation strategies in the certain probability to avoid being trapped in a local problem space.
    [Show full text]
  • Arxiv:1904.04123V2 [Stat.ML] 10 Oct 2019 Human Efforts Invested in finding Good Network Architec- Pected final Architecture, and Choose the Components That Tures
    ASAP: Architecture Search, Anneal and Prune Asaf Noy1, Niv Nayman1, Tal Ridnik1, Nadav Zamir1, Sivan Doveh2, Itamar Friedman1, Raja Giryes2, and Lihi Zelnik-Manor1 1DAMO Academy, Alibaba Group (fi[email protected]) 2School of Electrical Engineering, Tel-Aviv University Tel-Aviv, Israel Abstract methods, such as ENAS [34] and DARTS [12], reduced the search time to a few GPU days, while not compromising Automatic methods for Neural Architecture Search much on accuracy. While this is a major advancement, there (NAS) have been shown to produce state-of-the-art net- is still a need to speed up the process to make the automatic work models. Yet, their main drawback is the computa- search affordable and applicable to more problems. tional complexity of the search process. As some primal In this paper we propose an approach that further re- methods optimized over a discrete search space, thousands duces the search time to a few hours, rather than days or of days of GPU were required for convergence. A recent weeks. The key idea that enables this speed-up is relaxing approach is based on constructing a differentiable search the discrete search space to be continuous, differentiable, space that enables gradient-based optimization, which re- and annealable. For continuity and differentiability, we fol- duces the search time to a few days. While successful, it low the approach in DARTS [12], which shows that a con- still includes some noncontinuous steps, e.g., the pruning of tinuous and differentiable search space allows for gradient- many weak connections at once. In this paper, we propose a based optimization, resulting in orders of magnitude fewer differentiable search space that allows the annealing of ar- computations in comparison with black-box search, e.g., chitecture weights, while gradually pruning inferior opera- of [48, 35].
    [Show full text]
  • Lecture 19 -- Simulated Annealing
    Simulated Annealing Biostatistics 615/815 Lecture 19 Scheduling z Need to pick a date for mid-term z Default date is December 20, 2006 z We could have it earlier… • For example, on December 12, 2006? z What do you prefer? So far … z “Greedy” optimization methods • Can get trapped at local minima • Outcome might depend on starting point z Examples: • Golden Search • Nelder-Mead Simplex Optimization • E-M Algorithm Today … z Simulated Annealing z Markov-Chain Monte-Carlo method z Designed to search for global minimum among many local minima The Problem z Most minimization strategies find the nearest local minimum z Standard strategy • Generate trial point based on current estimates • Evaluate function at proposed location • Accept new value if it improves solution The Solution z We need a strategy to find other minima z This means, we must sometimes select new points that do not improve solution z How? Annealing z One manner in which crystals are formed z Gradual cooling of liquid … • At high temperatures, molecules move freely • At low temperatures, molecules are "stuck" z If cooling is slow • Low energy, organized crystal lattice formed Simulated Annealing z Analogy with thermodynamics z Incorporate a temperature parameter into the minimization procedure z At high temperatures, explore parameter space z At lower temperatures, restrict exploration Markov Chain z Start with some sample • A set of mixture parameters z Propose a change • Edit mixture parameters in some way z Decide whether to accept change • Decision is based on relative probabilities
    [Show full text]
  • Simulated Annealing
    Appendix A Simulated Annealing A.1 Introduction Simulated annealing can be considered as an extension of local optimization meth- ods because: • This approach can be applied to criteria that are neither continuous nor con- tinuously differentiable. It is just necessary to be able to compute the value of the criterion for any feasible solution. Thus, the criterion may be given by any type of function or even by an algorithm that returns a numerical value starting from the values of the parameters that define a solution. • The variables can be of a qualitative nature, needing only to be able to derive from their “values” a quantitative value of the criterion. Indeed, simulated annealing easily applies to combinatorial optimization prob- lems. A problem P that belongs to the field of application of simulated annealing is expressed as any optimization problem: Find *∈ Ss such that )*( = Opt sfsf )( ∈Ss where: • S is the set of feasible solutions. A feasible solution is one that satisfies all the constraints. We will see some examples in the next section. • f is a “function” in the broadest sense, as explained above. This function is the criterion. • Opt refers either to minimization or to maximization, depending on the type of problem. 450 A Simulated Annealing A.2 Basic Requirements In order to apply simulated annealing, we must be able to: • Compute the value of the criterion for any feasible solution. • Define an initial feasible solution. • Derive a neighboring feasible solution from any current solution. The criterion depends on the problem to be solved. An initial feasible solution may be difficult to find.
    [Show full text]
  • Heuristics: Simulated Annealing and Genetic Algorithms
    Simulated annealing and genetic algorithms CE 377K Stephen D. Boyles Spring 2015 1 Introduction Different optimization problems must be solved in different ways. Even just for the simple examples developed at the start of this course, the way that one solves the shortest path problem, for instance, is very different from the way that one solves the transit frequency setting problem. Later in this course, we will learn examples of these types of specific solution methods. But first, we start with a broad class of solution methods which can be applied to nearly any optimization problem. In optimization, there is often a tradeoff between how widely applicable a solution method is, and how efficient or effective it is at solving specific problems. For any one specific problem, a tailor- made solution process is likely much faster than a generally-applicable method, but generating such a method requires more effort and specialized knowledge, and is less \rewarding" in the sense that the method can only be more narrowly applied. So, the first solution methods you will see in this course are very widely applicable. They are examples of heuristics, which are not guaranteed to find the global optimum solution, but tend to work reasonably well in practice. In practice, they only tend to be applied for very large or very complicated problems which cannot be solved exactly in a reasonable amount of time with our current knowledge. But until you learn the more specialized solution methods later in the class, you can also use these even for smaller or simpler optimization problems too.
    [Show full text]
  • Improving Simulated Annealing Through Derandomization
    Improving Simulated Annealing through Derandomization Mathieu Gerber∗ Luke Bornn Department of Statistics Harvard University We propose and study a version of simulated annealing (SA) on continu- ¯ ous state spaces based on (t; s)R-sequences. The parameter R 2 N regulates the degree of randomness of the input sequence, with the case R = 0 corres- ponding to IID uniform random numbers and the limiting case R = 1 to (t; s)-sequences. Our main result, obtained for rectangular domains, shows that the resulting optimization method, which we refer to as QMC-SA, con- verges almost surely to the global optimum of the objective function ' for any R 2 N. When ' is univariate, we are in addition able to show that the completely deterministic version of QMC-SA is convergent. A key property of these results is that they do not require objective-dependent conditions on the cooling schedule. As a corollary of our theoretical analysis, we provide a new almost sure convergence result for SA which shares this property under minimal assumptions on '. We further explain how our results in fact apply to a broader class of optimization methods including for example threshold accepting, for which to our knowledge no convergence results currently ex- ist. We finally illustrate the superiority of QMC-SA over SA algorithms in a numerical study. Keywords: Global optimization; Quasi-Monte Carlo; Randomized quasi- Monte Carlo; Simulated annealing; Threshold accepting 1. Introduction arXiv:1505.03173v4 [stat.CO] 5 Sep 2016 Simulated annealing (SA) belongs to the class of stochastic optimization techniques, a popular suite of tools to find the global optimum of multi-modal and/or non-differentiable functions defined on continuous sets (Geman and Hwang, 1986).
    [Show full text]
  • Improved Salp Swarm Algorithm with Simulated Annealing for Solving Engineering Optimization Problems
    S S symmetry Article Improved Salp Swarm Algorithm with Simulated Annealing for Solving Engineering Optimization Problems Qing Duan, Lu Wang, Hongwei Kang * , Yong Shen, Xingping Sun and Qingyi Chen School of Software, Yunnan University, Kunming 650000, China; [email protected] (Q.D.); [email protected] (L.W.); [email protected] (Y.S.); [email protected] (X.S.); [email protected] (Q.C.) * Correspondence: [email protected] Abstract: Swarm-based algorithm can successfully avoid the local optimal constraints, thus achieving a smooth balance between exploration and exploitation. Salp swarm algorithm (SSA), as a swarm- based algorithm on account of the predation behavior of the salp, can solve complex daily life optimization problems in nature. SSA also has the problems of local stagnation and slow convergence rate. This paper introduces an improved salp swarm algorithm, which improve the SSA by using the chaotic sequence initialization strategy and symmetric adaptive population division. Moreover, a simulated annealing mechanism based on symmetric perturbation is introduced to enhance the local jumping ability of the algorithm. The improved algorithm is referred to SASSA. The CEC standard benchmark functions are used to evaluate the efficiency of the SASSA and the results demonstrate that the SASSA has better global search capability. SASSA is also applied to solve engineering optimization problems. The experimental results demonstrate that the exploratory and exploitative proclivities of the proposed algorithm and its convergence patterns are vividly improved. Citation: Duan, Q.; Wang, L.; Kang, H.; Shen, Y.; Sun, X.; Chen, Q. Keywords: swarm-based algorithm; salp swarm algorithm; single objective optimization; symmetric Improved Salp Swarm Algorithm perturbation; simulated annealing; engineering optimization problems with Simulated Annealing for Solving Engineering Optimization Problems.
    [Show full text]