A Schwartz Counterexample to the HRT Conjecture
This lightning talk presents an explicit twelve-term counterexample to the Heil–Ramanathan–Topiwala conjecture, proving that pairwise distinct time-frequency shifts of a nonzero Schwartz function can be linearly dependent. The presentation reveals the constructive proof architecture: exact dyadic coefficients, an irrational phase-space translation, a two-component vector Zak transform that overcomes topological obstructions, a certified numerical bound controlling a rank-one cocycle approximation, and a Diophantine Fourier solution to a scalar cohomological equation. The talk demonstrates how rigorous validated numerics, harmonic analysis, and dynamical systems combine to settle a long-standing open problem in time-frequency analysis.Script
For decades, mathematicians believed that time-frequency shifts of any nonzero function must be linearly independent. This paper shatters that conjecture with an explicit twelve-term counterexample built from a Schwartz function.
The construction specifies eleven phase-space points in a translated half-integer lattice plus the origin, exact dyadic coefficients, and a single Schwartz window whose twelve time-frequency translates sum exactly to zero despite each individual term having magnitude near one.
The scalar Zak transform creates a topological obstruction: continuous functions on the torus must vanish somewhere, and irrational translations propagate zeros everywhere. The authors bypass this by folding frequency into two components, producing a matrix-valued cocycle where rank-one invariant lines can survive.
The proof hinges on a certified numerical bound: the matrix field induced by the Weyl polynomial lies within operator distance one-third of an explicit rank-one model. This inequality was verified using 256-bit interval arithmetic over a grid of 2048 points, with rigorous derivative control between grid centers.
A contraction argument in the orthogonal complement constructs a smooth invariant vector field. The certified bound ensures the fiber map has Lipschitz constant below one, and iterative smoothness estimates using Faà di Bruno bounds prove the fixed point is infinitely differentiable, yielding a Schwartz-class eigenfunction.
The result transforms the HRT problem from a universal independence conjecture into a classification question: which configurations permit dependence, and what mechanisms are essential? The counterexample required an irrational translation, a translated lattice containing the origin externally, matrix-valued Zak methods, and Diophantine control. Explore the full construction and discover more breakthroughs in harmonic analysis at EmergentMind.com.
