simulated annealing -a informative approach
TRANSCRIPT
![Page 1: Simulated annealing -a informative approach](https://reader035.vdocument.in/reader035/viewer/2022062503/58e4b7db1a28ab1c1f8b5d13/html5/thumbnails/1.jpg)
Simulated Annealing
●Presented By:
Ranak Ghosh Septembar,2010
For more contact-08013835148
![Page 2: Simulated annealing -a informative approach](https://reader035.vdocument.in/reader035/viewer/2022062503/58e4b7db1a28ab1c1f8b5d13/html5/thumbnails/2.jpg)
CONTENT
●Motivation●The algorithm●Its applications●Examples●Conclusion
![Page 3: Simulated annealing -a informative approach](https://reader035.vdocument.in/reader035/viewer/2022062503/58e4b7db1a28ab1c1f8b5d13/html5/thumbnails/3.jpg)
Introduction
●Various algorithms proposed for placement in circuits.
●Constructive placement vs Iterative improvement.
●Simulated Annealing – an iterative improvement algorithm.
![Page 4: Simulated annealing -a informative approach](https://reader035.vdocument.in/reader035/viewer/2022062503/58e4b7db1a28ab1c1f8b5d13/html5/thumbnails/4.jpg)
Motivation
●Annealing in metals●Heat the solid state metal to a high
temperature●Cool it down very slowly according to a
specific schedule.●If the heating temperature is sufficiently high
to ensure random state and the cooling process is slow enough to ensure thermal equilibrium, then the atoms will place themselves in a pattern that corresponds to the global energy minimum of a perfect crystal.
Contd….
![Page 5: Simulated annealing -a informative approach](https://reader035.vdocument.in/reader035/viewer/2022062503/58e4b7db1a28ab1c1f8b5d13/html5/thumbnails/5.jpg)
Motivation●The connection between this algorithm and
mathematical minimization was first noted by Pincus.
●He proposed that it forms the basis of an optimization technique for combinatorial (and other) problems.
●SA's major advantage over other methods is an ability to avoid becoming trapped at local minima.
●The algorithm employs a random search which not only accepts changes that decrease objective function f, but also some changes that increase it.
●The latter is accepted with a probability P=exp(-df/T), where df is the increase in f and T is a control parameter,
![Page 6: Simulated annealing -a informative approach](https://reader035.vdocument.in/reader035/viewer/2022062503/58e4b7db1a28ab1c1f8b5d13/html5/thumbnails/6.jpg)
Simulated Annealing-The Algorithm
●Analogy
Cntnd.
![Page 7: Simulated annealing -a informative approach](https://reader035.vdocument.in/reader035/viewer/2022062503/58e4b7db1a28ab1c1f8b5d13/html5/thumbnails/7.jpg)
Analogy●Slowly cool down a heated solid, so that all particles
arrange in the ground energy state●At each temperature wait until the solid reaches its
thermal equilibrium●Probability of being in a state with energy E :
Pr { E = E } = 1/Z(T) . exp (-E / kB.T)
E EnergyT Temperature
kB Boltzmann constant
Z(T) Normalization factor (temperature dependant)
![Page 8: Simulated annealing -a informative approach](https://reader035.vdocument.in/reader035/viewer/2022062503/58e4b7db1a28ab1c1f8b5d13/html5/thumbnails/8.jpg)
PROCEDURE OF SIMULATED ANNEALING
![Page 9: Simulated annealing -a informative approach](https://reader035.vdocument.in/reader035/viewer/2022062503/58e4b7db1a28ab1c1f8b5d13/html5/thumbnails/9.jpg)
Cooling schedule
![Page 10: Simulated annealing -a informative approach](https://reader035.vdocument.in/reader035/viewer/2022062503/58e4b7db1a28ab1c1f8b5d13/html5/thumbnails/10.jpg)
Convergence of simulated annealing
![Page 11: Simulated annealing -a informative approach](https://reader035.vdocument.in/reader035/viewer/2022062503/58e4b7db1a28ab1c1f8b5d13/html5/thumbnails/11.jpg)
Ball on terrain example – Simulated Annealing vs Greedy Algorithms
●The ball is initially placed at a random position on the terrain. From the current position, the ball should be fired such that it can only move one step left or right.What algorithm should we follow for the ball to finally settle at the lowest point on the terrain?
![Page 12: Simulated annealing -a informative approach](https://reader035.vdocument.in/reader035/viewer/2022062503/58e4b7db1a28ab1c1f8b5d13/html5/thumbnails/12.jpg)
Ball on terrain example – SA vs Greedy Algorithms
![Page 13: Simulated annealing -a informative approach](https://reader035.vdocument.in/reader035/viewer/2022062503/58e4b7db1a28ab1c1f8b5d13/html5/thumbnails/13.jpg)
Example: Travel Salesperson Problem (TSP)
How to transverse n cities once and onlyonce with a minimal total distance?
![Page 14: Simulated annealing -a informative approach](https://reader035.vdocument.in/reader035/viewer/2022062503/58e4b7db1a28ab1c1f8b5d13/html5/thumbnails/14.jpg)
Traveling Salesperson Problem (TSP)
●TSP is famous discrete optimization problem●Many successful uses of SAN with TSP●Basic problem is to find best way for salesperson to hit
every city in territory once and only onceoSetting arises in many problems of optimization on
networks (communications, transportation, etc.)●If tour involves n cities, there are (n–1) ! /2 possible
solutionsoExtremely rapid growth in solution space as n
increasesoProblem is “NP hard”
●Perturbations in SAN steps based on three operations on network: inversion, translation, and switching
![Page 15: Simulated annealing -a informative approach](https://reader035.vdocument.in/reader035/viewer/2022062503/58e4b7db1a28ab1c1f8b5d13/html5/thumbnails/15.jpg)
9/8/2010
TSP●Move sets for TSP
1-2-3-4-5-6-7-8-9-10-11-12
1
2
3
4 5
67
8 9
10
11
12
1
2
3
4 5
67
8 9
10
11
12
1-2-3-4-5-9-8-7-6-10-11-12
1
2
3
4 5
67
89
10
11
12
1-2-3-4-8-7-5-9-6-10-11-12
1
2
3
4 5
67
89
10
11
12
1-2-11-4-8-7-5-9-6-10-3-12
Inversion
Switching
Translation
Inversion reverses order 6-7-8-9; translation removes section 6-7-8-9 and places 8-7 in place of 5-9; switching interchanges order of 3 and 11.
![Page 16: Simulated annealing -a informative approach](https://reader035.vdocument.in/reader035/viewer/2022062503/58e4b7db1a28ab1c1f8b5d13/html5/thumbnails/16.jpg)
Small-Scale TSP
●10-city tour (very small by industrial standards)oKnow by enumeration that minimum cost of tour =
440●Randomly chose inversion, translation, or switching
at each iterationoTuning required to choose “good” probabilities of
selecting these operators●8 of 10 SA runs find minimum cost tour
oSample mean cost of initial tour is 700; sample mean of final tour is 444
●Essential to success is adequate use of inversion operator; 0 of 10 SA runs find optimal tour if probability of inversion is 0.50
●SA successfully used in much larger TSPs oE.g., seminal 1983 (!) Kirkpatrick et al. paper in Science considers TSP with 400 cities
![Page 17: Simulated annealing -a informative approach](https://reader035.vdocument.in/reader035/viewer/2022062503/58e4b7db1a28ab1c1f8b5d13/html5/thumbnails/17.jpg)
Simulated Annealing for TSP-ANALOGY
●Initial configuration: permutation =>1, 2, 3, 4, …, Ntemperature T = 2cooling rate α = 0.99energy = d(1,2)+d(2,3)+…+d(N,1)●Generate a new configuration from the current one
at random●Evaluate ΔE = current energy – previous energy●If ΔE < 0 accept the current configuration (downhill)
Else accept configuration with probability :P=exp (-∆E / kB.T)●T= αT●Stopping criteria if energy < threshold or number of
iterations is reached.
![Page 18: Simulated annealing -a informative approach](https://reader035.vdocument.in/reader035/viewer/2022062503/58e4b7db1a28ab1c1f8b5d13/html5/thumbnails/18.jpg)
Case Study: 20-city problem
![Page 19: Simulated annealing -a informative approach](https://reader035.vdocument.in/reader035/viewer/2022062503/58e4b7db1a28ab1c1f8b5d13/html5/thumbnails/19.jpg)
Applications
●Circuit partitioning and placement.●Strategy scheduling for capital
products with complex product structure.
●Umpire scheduling in US Open Tennis tournament!
●Event-based learning situations.
![Page 20: Simulated annealing -a informative approach](https://reader035.vdocument.in/reader035/viewer/2022062503/58e4b7db1a28ab1c1f8b5d13/html5/thumbnails/20.jpg)
Jigsaw puzzles – Intuitive usage of Simulated Annealing
●Given a jigsaw puzzle such that one has to obtain the final shape using all pieces together.
●Starting with a random configuration, the human brain unconditionally chooses certain moves that tend to the solution.
●However, certain moves that may or may not lead to the solution are accepted or rejected with a certain small probability.
●The final shape is obtained as a result of a large number of iterations.
![Page 21: Simulated annealing -a informative approach](https://reader035.vdocument.in/reader035/viewer/2022062503/58e4b7db1a28ab1c1f8b5d13/html5/thumbnails/21.jpg)
Performance
●SA is a general solution method that is easily applicable to a large number of problems
●"Tuning" of the parameters (initial T, decrement of T, stop criterion) is relatively easy
●Generally the quality of the results of SA is good, although it can take a lot of time
●Results are generally not reproducible: another run can give a different result
●SA can leave an optimal solution and not find it again(so try to remember the best solution found so far)
●Proven to find the optimum under certain conditions; one of these conditions is that you must run forever
![Page 22: Simulated annealing -a informative approach](https://reader035.vdocument.in/reader035/viewer/2022062503/58e4b7db1a28ab1c1f8b5d13/html5/thumbnails/22.jpg)
Thank You!