Recurrent neural dynamics for handling linear equation system with rank-deficient coefficient and disturbance existence

A Novel Recurrent Neural Network Based Online Portfolio Analysis for High Frequency Trading

“…In Chen et al (2020) and Zhang et al (2020Zhang et al ( , 2022, researchers leverage deliberate design of activation functions for the speedup of convergence to finite time, which further enhances the real time processing capability of recurrent neural networks. Further investigations include the extension to complex-valued number dynamical signal processing (Xiao et al, 2021), noise-robust neural network derivation (Li et al, 2020;Xiao et al, 2022), model-free situations (Zhang et al, 2018), robot arm motion control (Li et al, 2018), mobile robot trajectory planning (Chen et al, 2021), multiple robot coordination (Li et al, 2017), Sylvester equation solving (Zhang et al, 2019), non-stationary quadratic programs (Qi et al, 2022) and rank-deficit problem solving (Shi et al, 2022), as comprehensively surveyed in Jin et al (2017).…”