Stability of neural networks and solitons of field theory
Sang-Hee Han, In Gyu Koh · Physical review. E, Statistical physics, plasmas, fluids, and related interdisciplinary topics · 1999
The layers of a feed-forward neural network are interpreted as a cascade of field theories. The stability of the neural network is interpreted as the topological stability of kink solutions. An explicit example is shown for a three-class problem with their field theoretical Lagrangian equations.