RILS-ROLS: robust symbolic regression via iterated local search and ordinary least squares

Journal of Big Data

Table 6 Comparison w.r.t. (\(R^2>0.999\)) solution accuracy (%)

Algorithm	All				Feynman				Strogatz
	Noise				Noise				Noise
	0	0.001	0.01	0.1	0	0.001	0.01	0.1	0	0.001	0.01	0.1
MRGP	92.69	91.54	88.46	1.92	93.1	92.24	91.81	2.16	89.29	85.71	60.71	0
Operon	86.92	86.54	86.54	73.46	86.21	85.78	85.78	82.33	92.86	92.86	92.86	0
RILS-ROLS	80	81.15	80	21.92	78.45	79.74	78.45	19.4	92.86	92.86	92.86	42.86
SBPGP	74.23	74.23	75	53.85	73.71	75.43	75	60.34	78.57	64.29	75	0
AI-Feynman	73.83	73.64	67.86	10.16	78.51	77.39	71.43	10.53	35.71	42.86	39.29	7.14
GPGOMEA	71.54	70.38	73.46	67.69	71.55	70.26	73.71	71.98	71.43	71.43	71.43	32.14
AFP-FE	55.77	50.38	50.38	50	59.05	52.16	52.59	53.45	28.57	35.71	32.14	21.43
EPLEX	44.23	45.38	52.31	46.92	46.98	47.84	56.03	51.72	21.43	25	21.43	7.14
AFP	42.69	41.92	40.38	40.77	44.83	44.4	42.67	43.97	25	21.43	21.43	14.29
FEAT	40	43.08	40.77	13.46	39.66	42.24	40.52	13.36	42.86	50	42.86	14.29
gplearn	30	29.23	27.13	21.92	32.76	31.9	29.13	23.71	7.14	7.14	10.71	7.14
ITEA	26.92	26.92	26.92	25.77	27.59	27.59	27.59	26.72	21.43	21.43	21.43	17.86
DSR	23.85	24.62	25	25	25	25.86	26.29	26.29	14.29	14.29	14.29	14.29
BSR	11.92	10.77	11.92	6.92	10.78	10.34	12.07	7.76	21.43	14.29	10.71	0
FFX	0	0	2.69	17.69	0	0	2.59	19.83	0	0	3.57	0