Table 16 Qualitative comparison of ELM models and classical neural networks
From: Extreme learning machine versus classical feedforward network
Criterion | BP | ELM |
|---|---|---|
# of hyperparameters | High | Low |
# of trainable parameters | Very high | Lower |
Training time | High | Low |
Memory requirements | Low | High |
Popularity | High | Low |
Supplementary techniques | Many | Few |
Community support | High | Low |
Big gamer’s interest | Very high | Low |
Model’s implementations | Numerous, up-to-date | Few, poorly supported |