Statement and proposed disproof
Conjecture 6.1 of Leonid V. Kovalev, Hardy-type norms of matrices, arXiv:2607.17373v1, printed page 9, proposes
for every real matrix of order . The definitions are
where is uniformly distributed on the Euclidean unit sphere. The conjectured constant, determined by a rank-two projection in the source, is
We give positive diagonal counterexamples to the first inequality in (1) in every order at least 500001. This refutes the full conjecture. The second inequality in (1) is not decided here. The dimension threshold is a convenient uniform estimate, not a claim about the first possible counterexample.
All comparisons in the proof are analytic or rational. The supplementary standard-library script verify.py repeats the finite rational arithmetic; it does not numerically integrate, search for matrices, or test a finite list of dimensions.
1. A Gaussian description of a dual unit vector
Let be a standard real Gaussian vector, and set
Gaussian radius and direction are independent, so
For any positive diagonal matrix , define
For every real matrix , pointwise Cauchy–Schwarz gives
In the middle equality, sign symmetry makes every off-diagonal expectation zero. All expectations are finite, because the absolute value of the inner product is bounded by . Equality holds when . This proves the needed supporting-functional identity for arbitrary matrices, without restricting the dual supremum to diagonal matrices.
Using (3), with equality attained at , yields
Consequently
For a diagonal vector , the fourth spherical moment identity is
For completeness, independent Gaussian moments give ; dividing by proves (5). Formula (2) implies
Combining this with (4) and (5) gives the exact identity
Thus it suffices to construct for which .
2. One large coordinate and many small coordinates
Fix
For every positive integer , take and
Let
By permutation symmetry of the last coordinates, the gradient vector is
Therefore
Introduce two fixed one-dimensional expectations, with standard normal:
Since , convexity of and conditional Jensen give
For every nonnegative and positive ,
If , this follows by decreasing the denominator; if , squaring reduces it to . Independence of and gives
Moreover, centering and Cauchy–Schwarz yield
Hence, for every ,
No limit in the matrix size has been used. Estimates (9) and (11) hold for each integer in this entire range.
3. Rational lower bounds for the two expectations
We prove
Put
where is Euler's constant. The elementary constant estimates proved in §5 are
In the integrals in (8), substitute on the positive half-axis. The standard integral
from DLMF 10.32.9, after also replacing by the integration variable, gives
The positive-axis expansions DLMF 10.25.2 and 10.31.2 are
where . All terms relevant here are positive. Since , (15) gives . Using and (13) in (14) yields
To bound the difference in (14), use and . Termwise differentiation of the convergent series gives
Before evaluating at , regard the coefficient in (15) as the function . Thus . Using (DLMF 10.29.3), differentiating (15) gives
Consequently this difference is greater than the positive rational number
Substitution of the rational values of gives
Equations (14), (18), and (19) prove the other part of (12). Both rational comparisons (16) and (19) are explicitly reproduced in verify.py.
4. Completion of the counterfamily
Set
For every , equations (7), (9), (11), and (12) imply
The difference between the last rational number and is exactly
It follows from (13) that . The positive diagonal matrix
therefore satisfies
by (4) and (6). These are full-rank real matrices. This proves the claimed disproof of (1).
5. Elementary certificates for the constants
Here are explicit finite bounds proving (13), so no decimal approximation of a special function is a premise.
For , define
The positive logarithmic series gives
Take ten terms at . The identities
then produce rational enclosures. Direct substitution in (22) shows
To enclose Euler's constant without a tabulated value, write
Each summand is between and . A telescoping upper sum and an integral lower estimate give
At , use the rational enclosure of from (22) and the exact finite sum . The resulting rational inequalities are
Subtracting (25) from (23) proves the two stated bounds for .
Finally, Machin's identity
follows from the tangent addition formula, with the resulting angle lying in the first quadrant. Enclose each arctangent by ten alternating series terms and its first omitted term. The resulting rational upper bound is less than . Squaring positive numbers then gives
which proves the remaining part of (13). The exact checker reproduces (23), (25), this comparison, (16), (19), and the final positive fraction (21).