Table 8 Comparison of the average RMSE and training times (in seconds) for ELM and MLP

Auto MPG

0.0784

0.0746

0.0010

0.9232

Bank

0.0474

0.0796

0.0028

1.4651

California Housing

0.1456

0.1494

0.0072

5.5622

Delta Ailerons

0.0394

0.0716

0.0012

1.1028

Machine CPU

0.0409

0.0813

0.0003

0.6753

Servo

0.1247

0.0795

0.0006

0.2217

Linear function (1 dim.)

0.0107

0.0078

0.0004

2.1336

Linear function (3 d.)

0.0021

0.0012

0.0005

1.0053

Linear function (10 d.)

0.0037

0.0018

0.0012

1.6946

Linear function (30 d.)

0.0056

0.0046

0.0040

2.2008

Linear function (100 d.)

0.0099

0.0050

0.0317

1.4333

Linear function (300 d.)

0.0216

0.0063

0.1311

6.3831

Linear function (1000 d.)

0.0366

0.0053

0.4212

138.9691

2nd deg. polynomial (1 d.)

0.1214

0.0078

0.0003

3.0669

2nd deg. polynomial (3 d.)

0.0721

0.0061

0.0004

6.1652

2nd deg. polynomial (10 d.)

0.0872

0.0073

0.0010

7.1019

2nd deg. polynomial (30 d.)

0.1202

0.1056

0.0042

4.7240

2nd deg. polynomial (100 d.)

0.2132

0.0010

0.0312

0.6630

2nd deg. polynomial (300 d.)

0.2600

0.1011

0.1291

2.9807

2nd deg. polynomial (1000 d.)

0.2297

0.0544

0.4217

44.3745

Sin

0.0570

0.2078

0.0010

3.1900

Cos

0.0522

0.2314

0.0013

4.5293

SinC

0.0608

0.1303

0.0013

0.7917

Friedman I (5 d.)

0.0596

0.0423

0.0014

4.5284

Friedman I (30 d.)

0.0963

0.0813

0.0047

1.5832

Friedman I (100 d.)

0.1326

0.0974

0.0330

1.2921

Friedman I (300 d.)

0.1651

0.1073

0.1353

7.2085

Friedman I (1000 d.)

0.1443

0.1401

0.4193

107.4185

Friedman II

0.0165

0.0405

0.0011

7.6029

Friedman III

0.0768

0.0949

0.0010

1.1100

Number of outcomes with results significantly better than the competing model’s

8

14

30

0