How statement and proof provenance work
The first chip identifies the source of the statement or construction; the second identifies the source of its local proof or verification.
- Literature-sourced: the exact statement appears in a cited source; only wording and notation differ.
- AI-adapted: a semantically identical restatement of literature-sourced material, modulo indexing, notation, and boundary cases adopted by the library.
- AI-generated: a genuinely novel statement formulated by AI, with no source for the claim itself.
These labels describe origin, not correctness: citations and verification chips remain separate evidence.
Smooth composition, difference, coefficient product and commutator estimates on bounded H^s data
Statement
Assume Countable Choice (The Axiom of Countable Choice ()). Let be an integer, let be open, let be an open time interval, let be a compact interval, and let and be smooth. Fix compact and . Constants below depend on these data and are uniform in . For with values in and norms at most :
(a) for .
(b) . The same bound holds at order zero.
(c) For , . For , . Commutators are interpreted distributionally by their Leibniz expansion, even if the two individual order- terms are not separately in .
(d) For , .
(e) Let be an integer, and with values in and . Then . Moreover for and , The same estimates apply to any smooth coefficient obtained from by differentiation or by a smooth matrix inverse where that inverse exists on a neighborhood of .
Facts & Assumptions
Given: Countable Choice, the smooth coefficient maps, compact time and value sets, and the stated norm bounds.
controls for integers (Periodic Sobolev embedding: H^s on the three-torus controls C^m when s minus m exceeds three halves).
Fourier truncation gives smooth density in every (The Galerkin projection is self-adjoint, commutes with derivatives, is a contraction in every H^r, converges to the identity and has the sharp Fourier-tail bound); Smooth maps are totally differentiable with their partial-derivative matrices (If all partial derivatives exist on a neighbourhood and are continuous at a point, then the map is totally differentiable there with Jacobian derivative), and (The chain rule for total derivatives: ). The one-variable Leibniz rule holds at every order (The general Leibniz rule for the -th derivative of a product). Applying it successively in each coordinate gives : each coordinate splits its derivatives independently, multiplying the binomial coefficients. Iterating the first-order chain rule gives the finite chain expansions used below, with the sum of orders on the field factors at most .
Proof
(Localize coefficients without convexity.) Cover by finitely many balls whose doubled closed balls lie in . In each doubled ball choose a smooth bump equal to on the inner ball and zero outside the doubled ball; such a radial bump is constructed from for and otherwise, whose derivatives at zero vanish since exponential decay dominates every polynomial. If these bumps are , put . Then on a neighborhood of and has compact support in . Extend and by zero in the value variable outside . They are smooth and equal to the original coefficients near . On times the torus times the compact convex hull of any fixed compact neighborhood of , all their required derivatives are bounded. This permits line-segment arguments even when is nonconvex.
(Composition (a).) For a smooth , the order- chain expansion consists of bounded coefficient derivatives times products with total order at most . At most one factor has order at least , since ; all others have order at most and are bounded in by [L1]. The possible exceptional factor has norm bounded by ; if there is no exceptional factor, use finite volume. Thus each term is at most , proving the asserted bound. For a general Sobolev field, Fourier truncations have bounded norms and converge uniformly to , with values eventually in a fixed compact neighborhood of . Apply the smooth bound to the extension of step 1.1. Fourier coefficients of the compositions converge by uniform convergence; finite weighted Fourier sums give the same bound for the limit. This also identifies the weak derivative expansion by testing the smooth approximations.
(Differences (b).) With the extended coefficient and , the pointwise identity holds. The maps have norm at most and take values in the compact convex hull of . Step 2.1 applied to the extended derivative bounds in uniformly. The algebra estimate [L2] gives the difference bound after integration. At order zero bounded gives the pointwise Lipschitz estimate and hence its version.
(Homogeneous commutators (c).) Expand . For , step 2.1 and [L1] bound the first factor in ; the second has at most derivatives. For the only remaining case is , where the first factor is bounded and is bounded by its norm. For and , at most one derivative falls on , which is bounded by its norm. These cases give precisely (c), without an additive constant. Density gives the distributional identity and bounds for general .
(Multipliers (d).) In the Leibniz expansion at order , coefficient derivatives at most are bounded. The only remaining case is a coefficient derivative of order , with no derivative on and ; its norm is bounded and has the required bound. Summation gives (d). Smooth approximation proves the claimed product membership for general Sobolev .
(Lower-norm-controlled high estimates (e).) With , derivatives of of order at most are bounded. In the order- chain expansion at most one factor has order at least , since . Its norm is bounded by unless its order is , in which case it is bounded by . This proves the high composition bound. In a commutator with , a coefficient derivative of order at most is bounded in , and the remaining derivative of has order at most . A coefficient derivative of order at least has bound , and at most two derivatives land on ; their bound follows from , since . Sum the finite expansion. Density extends the bound as in step 2.1. All smooth derived coefficients are covered by the same argument. This proves (e).
Depends on
The 2 pi-periodic three-torus, its normalized measure, test functions and the Sobolev spaces H^s
Periodic Fourier differentiation identity and equivalence of the derivative and Fourier-weight Sobolev norms
Periodic Sobolev embedding: H^s on the three-torus controls C^m when s minus m exceeds three halves
The periodic Sobolev space H^k of the three-torus is a Banach algebra for k at least four
maps and multi-index derivative notation in Euclidean space
The Axiom of Countable Choice ()
The Galerkin projection is self-adjoint, commutes with derivatives, is a contraction in every H^r, converges to the identity and has the sharp Fourier-tail bound
The chain rule for total derivatives:
If all partial derivatives exist on a neighbourhood and are continuous at a point, then the map is totally differentiable there with Jacobian derivative
The general Leibniz rule for the -th derivative of a product
Dependency tree · 2 levels
Sources
- H. Friedrich and A. D. Rendall, The Cauchy Problem for the Einstein Equations, Living Reviews in Relativity 3 (2000), article 1 (arXiv:gr-qc/0002074), Section 3.1 (Symmetric Hyperbolic Systems): the definition of a quasilinear symmetric hyperbolic system, the finite-dimensional mollifier approximation (3.1) solved by ODE theory, the energy and Moser product/composition estimates, the higher-order energy inequality, uniqueness, the Rellich/Ascoli compactness step and the passage to a classical solution, with Remarks 1-6 on smooth persistence, the continuation criterion, Cauchy stability and the non-identity time matrix
- J. K. Hunter, Notes on Partial Differential Equations (UC Davis lecture notes), Chapter 3 (Sobolev Spaces), Sections 3.1 (weak derivatives), 3.4 (properties of weak derivatives), 3.5 (Sobolev spaces), 3.7-3.8 (Sobolev embedding) and 3.10 (compactness results)