Parametric Stability Criteria for Delayed Recurrent Neural Networks via Flexible Delay-Dividing Method.
basic_science · Level V
Where this comes from
- Record sourced from PubMed, PMID 38865227.
- Also identified by DOI 10.1109/TNNLS.2024.3405964.
- No licence information is recorded for this record.
- Because redistribution is not established, this page shows the abstract only. Follow the links below for the full text.
Abstract
This article focuses on investigating the stability issue for recurrent neural networks (RNNs) with interval time-varying delays (TVDs) based on a flexible delay-dividing method with parameters, which are related to the delay derivative. First, an interval of delay is separated into parametric subintervals via the linear combination technique. Then, an establishment of Lyapunov-Krasovskii functional (LKF) is connected to the parameters, and a novel linear technology is suggested to dispose of integral terms in the derivatives of the constructed function. Finally, the validity and advantage of the inferred criteria are interpreted by the comparison of representative simulation examples.