An improved stability criterion of neural networks with time-varying delays in the form of quadratic function using novel geometry-based conditions
作者：
Highlights：
• The development of a novel negative conditions for a quadratic function based on geometry which could be optimal by adjusting a tuning parameter.
• The establishment of new Lyapunov functionals which are delay-product-type and lead the LMI conditions with high dependency on a quadratic function.
• The use of single integration of the neural activation function.
摘要
•The development of a novel negative conditions for a quadratic function based on geometry which could be optimal by adjusting a tuning parameter.•The establishment of new Lyapunov functionals which are delay-product-type and lead the LMI conditions with high dependency on a quadratic function.•The use of single integration of the neural activation function.
论文关键词：Neural networks,Time-varying delay,Stability,Quadratic function
论文评审过程：Received 10 December 2020, Revised 18 February 2021, Accepted 19 March 2021, Available online 6 April 2021, Version of Record 6 April 2021.
论文官网地址：https://doi.org/10.1016/j.amc.2021.126226
