Effect of shapes of activation functions on predictability in the echo state network
arXiv:1905.09419v17 citations
Originality Synthesis-oriented
AI Analysis
This work addresses the problem of optimizing activation functions for improved time series prediction in Echo State Networks, but it appears incremental as it focuses on comparing existing function types.
The study examined how different activation functions affect prediction accuracy in Echo State Networks for time series data, finding that certain functions with appropriate nonlinearity outperform the conventional sigmoid function.
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.