Dynamical properties of a perceptron learning process: structural stability under numerics and shadowing
DOI10.1007/s00332-011-9094-1zbMath1233.37061WikidataQ123956589 ScholiaQ123956589MaRDI QIDQ652279
Jerzy Ombach, Andrzej Bielecki
Publication date: 14 December 2011
Published in: Journal of Nonlinear Science (Search for Journal in Brave)
Full work available at URL: https://doi.org/10.1007/s00332-011-9094-1
robustness; dynamical system; topological conjugacy; Runge-Kutta methods; shadowing; inverse shadowing; gradient differential equation; perceptron learning process
68T05: Learning and adaptive systems in artificial intelligence
92B20: Neural networks for/in biological studies, artificial life and related topics
37C15: Topological and differentiable equivalence, conjugacy, moduli, classification of dynamical systems
65L20: Stability and convergence of numerical methods for ordinary differential equations
65L06: Multistep, Runge-Kutta and extrapolation methods for ordinary differential equations
37C50: Approximate trajectories (pseudotrajectories, shadowing, etc.) in smooth dynamics
37N30: Dynamical systems in numerical analysis
37M99: Approximation methods and numerical treatment of dynamical systems
Related Items
Cites Work
- Unnamed Item
- Unnamed Item
- Unnamed Item
- Shadowing property in analysis of neural networks dynamics.
- Deterministic convergence of an online gradient method for neural networks
- Shadowing in dynamical systems
- Topological conjugacy of discrete time-map and Euler discrete dynamical systems generated by a gradient flow on a two-dimensional compact manifold
- Structural Stability for the Euler Method
- Structural Stability of Morse--Smale Gradient-like Flows under Discretizations
- Hyperbolic homeomorphisms and bishadowing
- Estimation of the Euler method error on a Riemannian manifold
- Dynamical properties of learning process of weakly nonlinear and nonlinear neurons