The Setting
Take a finite index set , real node coordinates on , and three complex matrices weighted by discrete Gaussian masses of step size and variances centered at real triples summing to a level . Assembling those weighted matrices into a block matrix yields a matrix energy at each of the three vertex rows, and the problem is to bound the plane integral of their sum by the scaled cubic sum of entry moduli. Alongside this energy estimate lie uniform bounds on discrete Gaussian weight sums over grids of spacing when variances are at least , as well as monotonicity of the cubic entry sum under entrywise shrinkage.
Matrix Energy and Cubic Array Norms
The triangle vertex set is the set , equipped with cyclic addition and (written in the paper).✦ The triangle block space over an index set is the Cartesian product . Given an index set and a family of complex matrices, the triangle matrix on is the block matrix (1) whose triangle matrix block is for , for , and for : (1)
For an index set and a vertex , the vertex block projection on is the diagonal complex matrix whose diagonal entry at is if and otherwise.✦ For a finite index set , a family of complex matrices, and , the vertex star row on is the complex matrix .✦ The vertex star Gram matrix of a family of complex matrices at , for a finite index set , is the complex matrix on . The real trace of a complex matrix on a finite index set is the real part of the trace of .✦ Given a finite index set and a complex matrix , the matrix energy is the real number given by (2), where is the square root of : (2)✦
For an index set , a family of complex matrices, and a family of real functions , the weighted matrix array is the family of complex matrices defined entrywise by (3): (3)✦
At scale , the cubic array norm of a family of complex matrices on a finite index set is the real number given by (4): (4)✦ For an index set , a family of complex matrices, and a complex matrix , the zero-component replacement is the family of complex matrices indexed by equal to at and to at .✦
Gaussian Weights and Plane Integrals
The variance- Gaussian density for real and is the real value given by (5): (5)✦ The Gaussian weight for real , , , and is the real number .✦ Given an index set , , node coordinates , centers , and variances , the vertex heat weights form the family of real functions on defined by (6): (6) The mesh mass constant is the real number defined by (7): (7)✦
For an index set , a family of complex matrices, , , and , the heat matrix array is the weighted matrix array formed from and the vertex heat weights of (6).✦ For a finite index set , a family of complex matrices, , , , and , the heat matrix energy is the matrix energy of the vertex star Gram matrix of at , given by (8): (8)✦
At sum level , the constant-sum center plane is the subset of . Given , a sum level , and , the plane coordinate chart value is the triple in whose entries sum to , equal to at coordinate and to and at the remaining two vertices in increasing order.✦ The plane integral of a function at sum level is the Lebesgue integral over given by (9): (9)✦ The total plane heat energy of a family of complex matrices on a finite index set , with , , , and , is the real number defined by (10): (10)✦
The Claim
Theorem 1. Let , , , and let be an integer. Then the mesh sum satisfies (11): (11)✦ Given a finite subset , a step size , a variance , and a center , the weight sum over obeys (12): (12)✦ When is a finite index set, is a family of complex matrices, , , , for every , and satisfies for every center triple and every with vertex heat weights determined by , then (13) holds: (13)✦ When is finite, is a family of complex matrices, , , for every , and for , the total plane heat energy obeys (14): (14)✦ Take a finite index set , any , and two families and of complex matrices satisfying for all and ; then at scale the cubic array norms satisfy (15): (15)✦ If is finite, is a family of complex matrices, is a complex matrix with for all , and , then at scale the zero-component replacement obeys (16): (16)✦
In the paper, the result reads: for , an integer , finite complex arrays with entries , any , and variances , one has with (where stands for the heat matrix energy and for the measure on ), and the same bound holds if any entries are replaced by smaller entries in absolute value before forming the weighted matrices. In the stated theorem, (14) uses in place of integration against and in place of on arbitrary finite subsets , while (15) and (16) state entrywise shrinkage directly for .
What Is Checked
The claims (11) through (16) are stated over finite index sets and finite subsets , complex matrix families indexed by , shifts , integers , centers , node coordinates , sum levels , the Gaussian density and weights and , the mesh mass constant , the total plane heat energy , the cubic array norm , and zero-component replacement, under the hypotheses with or and a uniform bound on the weight sums, arbitrary in the cubic comparisons, and entrywise modulus dominance.