Broad Echo State Network with Reservoir Pruning for Nonstationary Time Series Prediction

Table 4

Network model parameters of the US power load dataset.

Hyperparameter	BPESN	BESN	ESN	BLS
Number of neurons in mapping layer	1–15	1–15	N/A	1–15
Number of neurons in enhancement layer	1–20	1–20	N/A	1–20
Reservoir size	500–1000	500–1000	500–1000	N/A
Spectral radius rate	0.95	0.95	0.95	N/A
Leaking rate	0.1	0.1	0.1	N/A
Sparseness	0.05	0.05	0.05	N/A
Pruning the number of times	1–10	N/A	N/A	N/A