Simulated Annealing

Komarudin
[email protected]
Random Search

2/28
 Hill climbing
Simple Iterative Improvement or Hill Climbing:
• Candidate is always and only accepted if cost
is lower (or fitness is higher) than current
configuration
• Stop when no neighbor with lower cost
(higher fitness) can be found

For minimization: steepest descent

Hill climbing
Steepest descent

5/28
 Hill climbing
Disadvantages:
• Local optimum as best result
• Local optimum depends on initial
configuration
• Generally no upper bound on iteration length
 How to cope with disadvantages
• Repeat algorithm many times with different initial
configurations
• Use information gathered in previous runs (not
preferable, Tabu Search)
• Use a more complex Generation Function to jump
out of local optimum (crossover mutation) 
Genetic Algorithm
• Use a more complex Evaluation Criterion that
accepts sometimes (randomly) also solutions away
from the (local) optimum
 Annealing process
• 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.
 Simulated Annealing
Use a more complex Evaluation Function:
• Do sometimes accept candidates with higher
cost to escape from local optimum
• Adapt the parameters of this Evaluation
Function during execution
• Based upon the analogy with the simulation of
the annealing of solids
Simulation of cooling (Metropolis 1953)
• At a fixed temperature T :
• Perturb (randomly) the current state to a new state
• E is the difference in energy between current and new state
• If E < 0 (new state is lower), accept new state as current
state
• If E  0 , accept new state with probability
Pr (accepted) = exp (- E / kB.T)
• Eventually the systems evolves into thermal equilibrium at
temperature T ; then the formula mentioned before holds
• When equilibrium is reached, temperature T can be lowered
and the process can be repeated
 Simulated Annealing
Step 1: Initialize – Start with a random initial placement.
Initialize a very high “temperature”.
Step 2: Move – Perturb the placement through a defined
move.
Step 3: Calculate score – calculate the change in the score
due to the move made.
Step 4: Choose – Depending on the change in score, accept
or reject the move. The prob of acceptance depending on
the current “temperature”.
Step 5: Update and repeat– Update the temperature value
by lowering the temperature. Go back to Step 2.
The process is done until “Freezing Point” is reached.
 Simulated Annealing
initialize;
REPEAT
REPEAT
perturb ( config.i  config.j, Cij);
IF Cij < 0 THEN accept
ELSE IF exp(-Cij/T) > random[0,1) THEN accept;
IF accept THEN update(config.j);
UNTIL equilibrium is approached sufficient closely;
T := next_lower(T);
UNTIL system is frozen or stop criterion is reached
 Analysis of SA
• Choose the start value of T so that in the beginning nearly all
perturbations are accepted (exploration), but not too big to
avoid long run times
• The function next_lower in the homogeneous variant is
generally a simple function to decrease T, e.g. a fixed part
(80%) of current T
• At the end T is so small that only a very small number of the
perturbations is accepted (exploitation)

• If possible, always try to remember explicitly the best solution

found so far; the algorithm itself can leave its best solution and
not find it again
 SA for TSP
• INIT-TEMP = 4000000;
• INIT-TSP = Random (can use heuristics);
• PERTURB(TSP)
1. 1-insert.
2. 1-swap.
3. 2-opt.
 Convergence of simulated annealing

AT INIT_TEMP Unconditional Acceptance

HILL CLIMBING Move accepted with

probability
= e-(^C/temp)
COST FUNCTION, C

HILL CLIMBING

HILL CLIMBING

AT FINAL_TEMP

NUMBER OF ITERATIONS
 Algorithm for TSP
Algorithm SA
Begin
t = t0;
cur_TSP = ini_TSP;
cur_score = SCORE(cur_TSP);
repeat
repeat
trial_TSP = {1-Insert, 1-Swap, 2-opt}(cur_TSP);
trial_score = SCORE(trial_TSP);
δs = trial_score – cur_score;
if (δs < 0) then
cur_score = trial_score;
cur_TSP = MOVE(comp1, comp2);
else
r = RANDOM(0,1);
if (r < e-(δs/t)) then
cur_score = trial_score;
cur_TSP = MOVE(comp1, comp2);
until (equilibrium at t is reached)
t = αt (0 < α < 1)
until (freezing point is reached)
End.
Ball on terrain example – SA vs Greedy Algorithms

Initial position
of the ball Simulated Annealing explores
more. Chooses this move with a
small probability (Hill Climbing)

Greedy Algorithm
gets stuck here!
Locally Optimum
Solution.

Upon a large no. of iterations,

SA converges to this solution.
 Run time
• Annealing: O*(n0.5) phases
• State-of-the-art walks [LV03]
– Worst case: O*(n) samples per phase
(for shape)
– O*(n3) steps per sample
• Total: O*(n4.5)
(compare to O*(n10) [GLS81], O*(n5) [BV02])
Terima kasih