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.
Each coefficient field of the modified logarithm is a holomorphic symplectic field with a global holomorphic primitive on the star-shaped domain, and
Statement
Let , let be analytic splitting data with , and let , , be the coefficient fields of The truncated binomial family , the truncated logarithm , its coefficient fields and the truncations (independent of the truncation order by Bounds for the truncated logarithm: and clause 4), with the bound of that lemma on the common retained domain . In this item denotes for every , including , whose coefficient is extracted using any order . Write in the coordinates of Complex phase domains, holomorphic maps on , and the complexified canonical form and bracket, and let be its complex Jacobian. Then for every :
- is a holomorphic vector field on : all are holomorphic on and real-valued on .
- on ; equivalently, the contraction one-form is closed there.
- There is a holomorphic with , and real-valued on , and
- ; that is, for every holomorphic on the domain of , .
Facts & Assumptions
Given: The analytic splitting data; the fields with components ; the bounds on (Bounds for the truncated logarithm: and ); the matrix and the form (Complex phase domains, holomorphic maps on , and the complexified canonical form and bracket, A phase domain, its canonical one-form and its standard symplectic two-form).
Coordinate bracket matrix formula: for holomorphic , , and the bracket is -bilinear and antisymmetric with constant (Calculus of holomorphic maps: chain rule, Wirtinger derivatives, coefficient extraction in one variable, and locally uniform limits clause 2, Complex phase domains, holomorphic maps on , and the complexified canonical form and bracket, Rectangular matrix multiplication and the identity matrix , including zero-sized shapes).
Derivation of the bracket: for the coefficient operators of the logarithm, for holomorphic , and acts on as (The binomial family of the splitting pullback is multiplicative and bracket-preserving, and its logarithm is a derivation clause 3, The truncated binomial family , the truncated logarithm , its coefficient fields and the truncations ).
Realness of the extensions and their flows. By the distance-domain estimate in Complex phase domains, holomorphic maps on , and the complexified canonical form and bracket, is star-shaped about . Fix ; the set of real for which has a connected open interval containing . The function is real analytic on that interval, since it is the restriction of a holomorphic function to a complex line. For in a neighborhood of , because and is open, so as and is real-valued. To justify continuation, let be the set of points where vanishes on a neighborhood. It is nonempty and open; if converges to an interior point , every derivative is zero, so continuity of all derivatives gives for every , and the real-analytic Taylor expansion at makes zero near . Thus is also closed in the interval and equals the whole interval. This proves ; the identical argument applies to . The restrictions of these extensions to the open real domain are real-valued, so their real-coordinate partial derivatives are real; holomorphic differentiability identifies those derivatives with the complex coordinate derivatives. Hence and are real on the real locus.
For either field or , the Picard iterates in The flow of a bounded holomorphic field on a polydisc: existence in complex time, displacement bound, holomorphy, and the composition law starting from real remain real when is real: the initial iterate is real, and each next iterate is , an integral of real vectors by induction. Their limit is real because is closed in . Thus each local real-time flow preserves real points; uniqueness and the local gluing in that lemma preserve this property along every real continuation where the flow is defined. The three factors defining therefore preserve real points for real , as do all defined iterates .
Closed one-forms on star-shaped real domains have potentials, with the explicit radial potential for centre (Poincare's lemma on a star-shaped domain: every closed C1 field is exact, Star-shaped open subsets of Euclidean space); the complex domain is open and star-shaped about (Complex phase domains, holomorphic maps on , and the complexified canonical form and bracket), and the real and imaginary parts of a closed complex one-form with coefficients are closed real one-forms.
Holomorphy criterion: a complex-valued function on an open subset of viewed as a real domain with for all is holomorphic, since then is -linear; and the zero-normalized radial primitive is real-valued on the real part of its star-shaped domain whenever is real on real points and real tangent vectors, because its integrand is then real (Complex phase domains, holomorphic maps on , and the complexified canonical form and bracket, Calculus of holomorphic maps: chain rule, Wirtinger derivatives, coefficient extraction in one variable, and locally uniform limits clause 2, Directional derivatives and partial derivatives of a map , The total (Fréchet) derivative as the linear first-order approximation with remainder).
Proof
Holomorphy and realness of the components. By The truncated binomial family , the truncated logarithm , its coefficient fields and the truncations , is a coefficient in of the holomorphic function (for any ), so it is holomorphic in by Calculus of holomorphic maps: chain rule, Wirtinger derivatives, coefficient extraction in one variable, and locally uniform limits clause 3; this is clause 1's first assertion. For real and real , is real by [L3], so each , hence , is real-valued there, and its Taylor coefficient in the real variable at is real; this gives the realness assertion.
Closedness. Fix and apply the bracket derivation rule [L2] to the constant bracket : Writing and using [L1], -component of the field, and , so, using and for , the identity reads for all , that is . Equivalently, write , so that and ; then , and comparing block entries shows that is precisely the symmetry of the Jacobian , i.e. the closedness .
The first coefficient. Since and , the coefficient of in is ; here the derivative of at is : differentiating at and using and gives . Hence , which is clause 4.
Primitive. Let and be the real and imaginary parts of , viewed as real one-forms on the open star-shaped set ; both are closed because . By [L4] there are potentials with , ; subtracting constants we may assume . Set . Then , and because is a -form; hence is holomorphic by [L5]. At real points and for real tangent vectors , is real by the realness of the coefficients from step 1.1, so vanishes on real directions at real points; the real part of is star-shaped and connected, so there and is real-valued on . The gradient bound follows from and , and by the radial form of the potential and the norm inequality.
Conclusion. Clause 1 is step 1.1, clause 2 is step 1.2, clause 3 is step 2.1 and clause 4 is step 1.3.
Remarks
- The star-shapedness of (and hence of and every ) is used exactly here: it produces global holomorphic primitives on the working domain. On a domain with nonvanishing first de Rham cohomology the closed form would only be locally exact, and this is the step the design records as requiring the star-shaped hypothesis.
- The primitive is unique up to an additive constant on the connected domain ; the normalization fixes it. The modified Hamiltonian of Optimal finite truncation gives an exponential local splitting defect is built from these normalized primitives.
Depends on
The truncated binomial family , the truncated logarithm , its coefficient fields and the truncations
Bounds for the truncated logarithm: and
The binomial family of the splitting pullback is multiplicative and bracket-preserving, and its logarithm is a derivation
Hamiltonian flows and the exact splitting map preserve the complexified canonical form and the Poisson bracket
The flow of a bounded holomorphic field on a polydisc: existence in complex time, displacement bound, holomorphy, and the composition law
Calculus of holomorphic maps: chain rule, Wirtinger derivatives, coefficient extraction in one variable, and locally uniform limits
Iterated Cauchy formula, power-series expansion and Cauchy estimates for functions continuous and separately holomorphic on a polydisc
Analytic splitting data with uniform holomorphic neighbourhood, and the exact symmetric splitting map
Complex phase domains, holomorphic maps on , and the complexified canonical form and bracket
A phase domain, its canonical one-form and its standard symplectic two-form
Poincare's lemma on a star-shaped domain: every closed C1 field is exact
Star-shaped open subsets of Euclidean space
The chain rule for total derivatives:
The total (Fréchet) derivative as the linear first-order approximation with remainder
Directional derivatives and partial derivatives of a map
Rectangular matrix multiplication and the identity matrix , including zero-sized shapes
The transpose of a matrix
Heine-Borel in : with the Euclidean metric a subset of is compact if and only if it is closed and bounded, and the proof by bisection uses no choice principle; the same holds on the real line
Open ball, closed ball and sphere in a metric space
The metric topology: a set is open when every one of its points has a ball around it inside the set; closed means open complement
Dependency tree · 2 levels
Sources
- E. Hairer, Geometric Numerical Integration (TU Munchen 2010), Lecture 3 'Backward error analysis', Section 2: existence of a local modified Hamiltonian for symplectic methods and its global existence on star-shaped domains, printed pages 5-6
- E. Hairer, Geometric Numerical Integration (TU Munchen 2010), Lecture 1 'Hamiltonian systems', Section 3: symplectic transformations and the closedness of the contracted form, printed pages 5-7