Command line arguments example

AN=SA

Reference

S. Kirkpatrick, C. D. Gelatt Jr, and M. P. Vecchi. (1983). “Optimization by Simulated Annealing”. Science. 220 (4598): 671–680.

Pseudocode

1. Let $s=s_0$
2. For $k\in \{1, \ldots, k_{max}\}$:
   1. T $\gets$ temperature($\frac{k_{max}}{k}$)
   2. Pick a random neighbour, $s_{new}\gets$ neighbour($s$)
   3. If P(E($s$), E($s_{new}$), T) $\leq$ random(0, 1)
      1. $s \gets s_{new}$
3. Output the final state $s$

Class view

class simulated_annealing : public algorithm

Link: algorithm

Added data members

Name	Type	Utility
m_k_max	size_t	Maximum number of evalutions
m_s	solution<>	Candidate state (solution) $s$
m_s_new	solution<>	Neighboring state $s_{new}$

Member function

Name	Utility
initialize()	Initialize and evalute $s$
run_()	Optimize $s$ according to Pseudocode 2.
record()	Record the number of evaluations and the error (difference of the first objective between the current best solution and the global optima)

Added member functions

Name	Utility
temperature(val)	The function temperature in Pseudocode 2.1.
neighbour(s, s_new)	The function neighbour in Pseudocode 2.2.
P(s, s_new, T)	The acceptance probability function P in Pseudocode 2.3.