Adaptive classification of temporal signals in fixed-weights recurrent neural networks: an existence proof

2026-07-072026-07-07http://salesiana.dossiersoluciones.com/handle/123456789/129413We address the important theoretical question why a recurrent neural network with fixed weights can adaptively classify time-varied signals in the presence of additive noise and parametric perturbations. We provide a mathematical proof assuming that unknown parameters are allowed to enter the signal nonlinearly and the noise amplitude is sufficiently small.22 pagesOptimization and ControlDynamical Systems82C32; 35B40; 37C70; 68T05Adaptive classification of temporal signals in fixed-weights recurrent neural networks: an existence prooftext