[Retracted] Gradient Descent Optimization in Deep Learning Model Training Based on Multistage and Method Combination Strategy

Table 9

Performance of the proposed method, RMSprop.

	ResNet-20 on Cafri-10		LSTM on IMDB
	Val-loss	Val-acc	Val-loss	Val-acc
RMSprop + SGD	0.5968	0.8513	0.4859	0.8358
RMSprop + (SGD + d)	0.5889	0.8529	0.4676	0.8344
RMSprop + (SGD + M)	0.7347	0.8057	0.5672	0.8311
RMSprop + (SGD + M + d)	0.8832	0.7754	0.5583	0.8309
RMSprop + RMSprop	0.8139	0.7978	0.8128	0.8143
RMSprop + (RMSprop + d)	0.7960	0.7927	0.7715	0.8126
RMSprop + Adam	0.9790	0.7513	0.9848	0.8129
RMSprop + (Adam + d)	0.6724	0.8258	0.9556	0.8126

The bold values represent the best results.