Observable Quotients and Exact Projected Dynamics: Closure, Memory, Minimal Dynamical Completion, and Effective Field Operators
Nonlinear dynamical descent
Nonlinear dynamical descent
The criterion has an exact nonlinear counterpart, obtained by replacing the linear quotient map by a surjective submersion and the linear generator by a smooth vector field.
Assumption 8.1.
and are smooth manifolds, is a smooth surjective submersion (so is surjective for every , and is onto), and is a smooth vector field on .
The hypothesis that be a surjective submersion is not cosmetic. It guarantees (i) that is a genuine smooth manifold and an embedded submanifold for each , and (ii) the existence of smooth local sections, which is what makes the smoothness argument in Theorem 8.2 work. We make no claim about arbitrary set-theoretic quotients: a quotient of a manifold by an arbitrary equivalence relation need not be a smooth manifold at all, and nothing below should be read as asserting otherwise.
Theorem 8.2 (Nonlinear projectability criterion).
Under Assumption 8.1, there exists a smooth vector field on with if and only if applied to is constant along every fiber: When it exists, is unique.
Proof. Necessity. Suppose exists and . Then by (8.1), , and both sides lie in the same tangent space , so (8.2) holds.
Sufficiency. Assume (8.2). For , use surjectivity of to pick and define By (8.2) this is independent of the chosen , so is a well-defined section of : by construction. Equation (8.1) holds by definition.
Smoothness. Fix and . Because is a submersion, the local normal form for submersions [8, Thm. 4.12] provides an open neighbourhood in and a smooth local section with and . On we then have, from (8.3) applied with , which is a composition of smooth maps ( smooth, smooth as a section of , smooth as the tangent map of a smooth map), hence smooth on . Different choices of section give the same on the overlap by fiber constancy (8.2), so these local formulas patch to a globally well-defined smooth field. Since was arbitrary, is smooth on .
Uniqueness. If and both satisfy (8.1), then for any pick (possible by surjectivity); then . ◻
Remark 38 (The nonlinear analogue of ).
Failure of (8.2) says: there are two source states with the same readout whose projected velocities differ. This is verbatim the phenomenon of Theorem 6.4, where is the linear map , the fibers are the affine subspaces , and the projected velocity at is , which is fiber-constant precisely when . Thus the linear closure criterion is the linearization, and its exact generalization: the flow descends to the observable manifold exactly when equal present readouts force equal projected velocities. Nothing in Theorem 8.2 produces a memory kernel; in the nonlinear setting the exact reduction identity of Theorem 6.2 has no elementary counterpart, and the theorem should be read as supplying only the sharp criterion for descent, not a reduction formula.