The Setting
Take a bounded open interval with , a natural number , a parameter for a spectral exponent , and positive spectral parameters . For an -orthonormal family of real amplitudes on vanishing at and , how does maximizing an action functional over linear combinations with real lower triangular matrices control the integrated moments ?
These notes go through the finite-interval action inequality and its companion matrix-flow estimates, together with the definitions they are built on.
The Definitions
For real endpoints and , the Sobolev space consists of pairs of real-valued functions on that are square-integrable with respect to Lebesgue measure on , with continuous on , satisfying , and obeying the weak derivative identity (1)✦ for every smooth compactly supported test function whose support lies in .
For a natural number , the space of real lower triangular matrices is the set of real matrices satisfying for all indices and with .✦
Given , real numbers , a vector with diagonal matrix , a family whose components are pairs , and a real matrix , the finite-interval action is defined by (2)✦ where and .
With , , and vectors , the action density is (3)✦
For a natural number , the space of real symmetric matrices is the set of real matrices satisfying .✦
To each , vector , real matrix , and scalar corresponds the matrix pencil defined by (4)✦ where is the identity matrix.
For , the regularized scalar primitive is defined by (5)✦
The field normalization constant for is defined by (6)✦
Using (4), (5), and (6), the truncated matrix field for , , , and is defined by (7)✦ where denotes applied to via continuous functional calculus.
In the limit , the regularized matrix field for , , , and is defined from (7) by (8)✦
When and , for any and , the symmetric matrix field is the restriction of the regularized matrix field (8) to .✦
Associated with (8) for , , , and is the matrix field primitive , defined by (9)✦
On , for , , and , the action defect is given by (10)✦
The constraint defect for , , , , and is defined from (8) by (11)✦
The Claim
Theorem 1. Let , let be real endpoints, let , and let satisfy for every , with . If satisfies when and when for all indices and , then in the extended real numbers , (12)✦
In the paper, the result reads: let satisfy and let ; then , where the supremum is over all real lower triangular matrices, , and stands for .
Theorem 2. Let , let , let and be real numbers, and let satisfy for every . Suppose are continuous on with , each component is differentiable on with derivative , and for all , equals when and when . If for every , then (13)✦
Theorem 3. Let , let , let , let and be real numbers, and let satisfy for every , with . Suppose are continuous on with , each component is differentiable on with derivative , and for all , equals when and when . If for every , then (14)✦
Theorem 4. Let , let , let , let and be real numbers, and let with . Suppose are continuous on with and each component differentiable on with derivative , and suppose satisfies and , is differentiable within at every with derivative , and satisfies for every . Then (15)✦
What Is Checked
The action inequality (12) is stated over , , , positive vectors , orthonormal families in satisfying (1), lower triangular matrices in , and the finite-interval action (2); the upper-bound estimate (13) and the regularized bound (14) are stated over , , , positive vectors , pairs continuous on with on and whose components are orthonormal on , lower triangular matrices in , the action density (3), and, for (14), and the matrix field primitive (9); and the penalized action inequality (15) is stated over , , , , , , pairs continuous on with on and , symmetric matrix flows in with boundary values and derivative within driven by the symmetric matrix field and symmetric outer product, and the action and constraint defects (10) and (11).