Table 2.

The comparison of the performance when a constant learning rate and a decreasing learning rate are applied

	Constant Learning Rate		Decreasing Learning Rate
Input Parameters	Running Time	Accuracy (Average dSCC)	Running Time	Accuracy (Average dSCC)
CHR = 1-22, NUM_STR = 1, ALPHA = constant	4 min	0.821	13 s	0.8493
CHR = 1-22, NUM_STR = 1, ALPHA = [0.1, 2]	1 h, 30 min	0.8456	3 min	0.8536
CHR = 1-22, NUM_STR = 5, ALPHA = [0.1, 2]	7 h	0.8546	20 min	0.8546
CHR = 1, NUM_STR = 1, ALPHA = constant	37 s	0.7556	2 s	0.8088
CHR = 1, NUM_STR = 5, ALPHA = [0.1, 2]	1 h	0.7841	3 min	0.8088
CHR = 21, NUM_STR = 1, ALPHA = constant	0.7 s	0.8969	0.2 s	0.8995
CHR = 21, NUM_STR = 5, ALPHA = [0.1, 2]	36 s	0.9018	2 s	0.9018
CHR = 21, NUM_STR = 30, ALPHA = [0.1, 2]	4 min	0.9018	12 s	0.9018
CHR = 21, NUM_STR = 50, ALPHA = [0.1, 2]	6 min	0.9021	18 s	0.9018
CHR = 21, NUM_STR = 100, ALPHA = [0.1, 2]	12 min	0.9020	37 s	0.9020
CHR = 21, NUM_STR = 200, ALPHA = [0.1, 2]	24 min	0.9022	83 s	0.9020
CHR = 21, NUM_STR = 500, ALPHA = [0.1, 2]	1 h	0.9022	3 min	0.9021

The comparison of the computing time and the average dSCC value obtained by using a constant or a decreasing learning rate for different input parameters for the chromosome 1 – 22 of the GM06990 cell line. We used the constant learning rate 0.0001, and we defined the initial_λ = 0.01 for the decreasing learning rate. CHR represents the chromosome number, and NUM_STR represents the number of ensemble structures generated per conversion factor(α), ALPHA represents the conversion factor. The decreasing learning rate achieved a better computing speed in all the cases