Effect of shapes of activation functions on predictability in the echo state network

2019/05/22 by Hanten Chang, Shinji Nakaoka, Chang, Hanten +3 · 1 citation
Computer Science · #FOS: Computer and information sciences #Machine Learning (cs.LG) #Neural and Evolutionary Computing (cs.NE) #cs.LG #cs.NE

paper · pdf · doi:10.48550/arxiv.1905.09419

arxiv created 2019/05/22 · arxiv updated 2019/05/24

Abstract

We investigate prediction accuracy for time series of Echo state networks with respect to several kinds of activation functions. As a result, we found that some kinds of activation functions with an appropriate nonlinearity show high performance compared to the conventional sigmoid function.

Cited by

• Empirical Investigation into Configuring Echo State Networks for Representative Benchmark Problem Domains

Related

• Critical Echo State Networks that Anticipate Input using Morphable Transfer Functions
• Tailoring Artificial Neural Networks for Optimal Learning
• The Asymptotic Performance of Linear Echo State Neural Networks
• A local Echo State Property through the largest Lyapunov exponent
• Feature-Based Echo-State Networks: A Step Towards Interpretability and Minimalism in Reservoir Computer