{"id":"arxiv/2610.25023","url":"https://arxiv.org/abs/2610.25023","tags":["math.NA","cs.LG","math.SG"],"score":0,"title":"Integrable building blocks for symplectic neural networks: learned-type layers, fiber-preserving layers, and what a single layer can do","domain":"math.NA","source":"arxiv","authors":["Andrew Qing He"],"reviews":0,"subject":"math.NA","summary":"Exactly symplectic neural networks are built almost exclusively from shear maps and gain expressivity only through depth. We ask what else a single explicit symplectic layer can be. Observing that every layer in use is the exact flow of a Hamiltonian that is a function of commuting integrals, we obtain three families of layers: a learned-type layer, in which every conjugate pair is rotated, squeezed or sheared by a state-dependent amount and the type is found by training; moving-plane shears, residual symplectic maps outside the known ridge-function class; and fiber-preserving layers, which…","category":"research","comments":0,"disputed":false,"division":"math","mentions":0,"discussed":0,"published":"2026-10-18","reactions":{}}