Adaptive Cuckoo Search Algorithm for Unconstrained Optimization

Table 2

Performance comparison of ACSA with other optimization methods (in terms of fixed iteration number).

Benchmark function	Generation	Average best fitness value over 30 independent runs
		DE	EP	GA	PSO	SA	CSA	ACSA
Ackley	1	19.15	19.62	20.93	20.77	21.08	20.95	20.90
	500	2.68	4.94	11.54	2.69	20.54	11.16	3.76
	1000	2.68	4.63	8.44	2.69	19.43	4.01	0.06
	1500	2.68	4.41	6.94	2.69	17.57	1.90
	2000	2.68	4.29	6.00	2.69	15.39	0.50
de Jong	1	32.88	41.85	334.35	334.12	332.88	335.59	343.79
	500	0.02	1.44	17.79	0.27	89.98	1.91	0.14
	1000	0.02	1.21	7.65	0.27	68.02	0.02
	1500	0.02	1.13	4.52	0.27	40.57
	2000	0.02	1.04	2.96	0.27	20.32
Griewank	1	96.76	155.82	2521.20	2477.39	289.59	2516.57	2543.72
	500	0.35	0.92	342.43	2.13	252.83	89.87	69.32
	1000	0.34	0.85	199.79	2.13	197.62	7.40	4.23
	1500	0.33	0.83	136.00	2.13	147.97	1.46	1.17
	2000	0.33	0.78	103.27	2.13	106.96	1.01	0.70
Rastrigin	1	119.64	125.49	122.83	126.52	191.39	123.00	122.26
	500	14.69	45.90	8.16	28.09	186.84	3.67	2.94
	1000	14.69	41.27	4.27	28.08	173.08	0.62	0.25
	1500	14.69	38.54	3.21	27.84	165.48
	2000	14.69	36.82	2.28	27.78	158.28
Rosenbrock	1
	500						205.31	61.01
	1000						13.82	6.37
	1500						4.72	3.17
	2000						2.94	1.75