A new parameterized activation function, S4, that blends sigmoid and softsign through a smooth sigmoid-weighted transition is claimed to improve accuracy and convergence on small neural network benchmarks.
https://www.kaggle.com/c/boston-housing
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
REJECT 1representative citing papers
citing papers explorer
-
Hybrid activation functions for deep neural networks: S3 and S4 -- a novel approach to gradient flow optimization
A new parameterized activation function, S4, that blends sigmoid and softsign through a smooth sigmoid-weighted transition is claimed to improve accuracy and convergence on small neural network benchmarks.