notesum.ai

Published at November 18

Recurrent Stochastic Configuration Networks with Incremental Blocks

cs.AI

Released Date: November 18, 2024

Authors: Gang Dang¹, Dainhui Wang

Aff.: ¹State Key Laboratory of Synthetical Automation for Process Industries Northeastern University, Shenyang 110819, China

Arxiv: http://arxiv.org/abs/2411.11303v1

Refer to caption

Datasets	Models	Reservoir size	Training time	Training NRMSE	Testing NRMSE
MG	ESN	96	0.14271±0.03564	0.01172±0.00208	0.02573±0.00596
	DESN	80	0.11035±0.05237	0.00634±0.00056	0.01722±0.00739
	GESN	90	0.28762±0.19001	0.00405±0.00062	0.01563±0.00851
	RSCN	68	0.95326±0.22615	0.00338±0.00045	0.01207±0.00619
	BRSCN	50	0.76889±0.14271	0.00316±0.00047	0.01119±0.00142
MG1	ESN	124	0.16983±0.05084	0.01572±0.00679	0.03983±0.01130
	DESN	100	0.12938±0.07361	0.01192±0.00102	0.02736±0.00572
	GESN	100	0.40426±0.15372	0.00673±0.00054	0.02218±0.00376
	RSCN	79	0.88756±0.53245	0.00509±0.00047	0.01433±0.00378
	BRSCN	70	0.83928±0.57240	0.00483±0.00069	0.01346±0.00221
MG2	ESN	135	0.15339±0.07362	0.03091±0.01927	0.08309±0.01122
	DESN	120	0.12982±0.04551	0.01647±0.00887	0.07028±0.00983
	GESN	110	0.33823±0.22938	0.01009±0.00763	0.05128±0.00436
	RSCN	105	1.21287±0.29201	0.00712±0.00550	0.03516±0.00249
	BRSCN	110	1.15741±0.28523	0.00728±0.00115	0.03129±0.00285