Echo State Property upon Noisy Driving Input
The echo state property (ESP) is a key concept for understanding the working principle of the most widely used reservoir computing model, the echo state network (ESN). The ESP is achieved most of the operation time under general conditions, yet the property is lost when a combination of driving inpu...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Wiley
2024-01-01
|
| Series: | Complexity |
| Online Access: | http://dx.doi.org/10.1155/2024/5593925 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849691206100975616 |
|---|---|
| author | Junhyuk Woo Hyeongmo Kim Soon Ho Kim Kyungreem Han |
| author_facet | Junhyuk Woo Hyeongmo Kim Soon Ho Kim Kyungreem Han |
| author_sort | Junhyuk Woo |
| collection | DOAJ |
| description | The echo state property (ESP) is a key concept for understanding the working principle of the most widely used reservoir computing model, the echo state network (ESN). The ESP is achieved most of the operation time under general conditions, yet the property is lost when a combination of driving input signals and intrinsic reservoir dynamics causes unfavorable conditions for forgetting the initial transient state. A widely used treatment, setting the spectral radius of the weight matrix below the unity, is not sufficient as it may not properly account for the nature of driving inputs. Here, we characterize how noisy driving inputs affect the dynamical properties of an ESN and the empirical evaluation of the ESP. The standard ESN with a hyperbolic tangent activation function is tested using the MNIST handwritten digit datasets at different additive white Gaussian noise levels. The correlations among the neurons, input mapping, and memory capacity of the reservoir nonlinearly decrease with the noise level. These trends agree with the deterioration of the MNIST classification accuracy against noise. In addition, the ESP index for noisy driving input is developed as a tool to help easily assess ESPs in practical applications. Bifurcation analysis explicates how the noise destroys an asymptotical convergence in an ESN and confirms that the proposed index successfully captures the ESP against noise. These results pave the way for developing noise-robust reservoir computing systems, which may promote the validity and utility of reservoir computing for real-world machine learning applications. |
| format | Article |
| id | doaj-art-e280d6f8c6a9441faf46826ca02ebf32 |
| institution | DOAJ |
| issn | 1099-0526 |
| language | English |
| publishDate | 2024-01-01 |
| publisher | Wiley |
| record_format | Article |
| series | Complexity |
| spelling | doaj-art-e280d6f8c6a9441faf46826ca02ebf322025-08-20T03:21:06ZengWileyComplexity1099-05262024-01-01202410.1155/2024/5593925Echo State Property upon Noisy Driving InputJunhyuk Woo0Hyeongmo Kim1Soon Ho Kim2Kyungreem Han3Laboratory of Computational NeurophysicsLaboratory of Computational NeurophysicsLaboratory of Computational NeurophysicsLaboratory of Computational NeurophysicsThe echo state property (ESP) is a key concept for understanding the working principle of the most widely used reservoir computing model, the echo state network (ESN). The ESP is achieved most of the operation time under general conditions, yet the property is lost when a combination of driving input signals and intrinsic reservoir dynamics causes unfavorable conditions for forgetting the initial transient state. A widely used treatment, setting the spectral radius of the weight matrix below the unity, is not sufficient as it may not properly account for the nature of driving inputs. Here, we characterize how noisy driving inputs affect the dynamical properties of an ESN and the empirical evaluation of the ESP. The standard ESN with a hyperbolic tangent activation function is tested using the MNIST handwritten digit datasets at different additive white Gaussian noise levels. The correlations among the neurons, input mapping, and memory capacity of the reservoir nonlinearly decrease with the noise level. These trends agree with the deterioration of the MNIST classification accuracy against noise. In addition, the ESP index for noisy driving input is developed as a tool to help easily assess ESPs in practical applications. Bifurcation analysis explicates how the noise destroys an asymptotical convergence in an ESN and confirms that the proposed index successfully captures the ESP against noise. These results pave the way for developing noise-robust reservoir computing systems, which may promote the validity and utility of reservoir computing for real-world machine learning applications.http://dx.doi.org/10.1155/2024/5593925 |
| spellingShingle | Junhyuk Woo Hyeongmo Kim Soon Ho Kim Kyungreem Han Echo State Property upon Noisy Driving Input Complexity |
| title | Echo State Property upon Noisy Driving Input |
| title_full | Echo State Property upon Noisy Driving Input |
| title_fullStr | Echo State Property upon Noisy Driving Input |
| title_full_unstemmed | Echo State Property upon Noisy Driving Input |
| title_short | Echo State Property upon Noisy Driving Input |
| title_sort | echo state property upon noisy driving input |
| url | http://dx.doi.org/10.1155/2024/5593925 |
| work_keys_str_mv | AT junhyukwoo echostatepropertyuponnoisydrivinginput AT hyeongmokim echostatepropertyuponnoisydrivinginput AT soonhokim echostatepropertyuponnoisydrivinginput AT kyungreemhan echostatepropertyuponnoisydrivinginput |