HRT Conjecture Partially Resolved with AI Assistance

Resolution of the HRT Conjecture

The HRT conjecture, which posits that a non-zero function cannot have finite linear relations between its time-frequency shifts, has been partially resolved. Specifically, the Schwartz case of the conjecture, which assumes the function is a Schwartz function, has been disproven. Researchers Faulhuber, Petersen, van Velthoven, and Voigtlaender demonstrated the existence of such a function.

The Counterexample

The new theorem establishes that there exist complex numbers, distinct points, and a non-zero Schwartz function that satisfy the previously conjectured impossible relation. This counterexample expands the parameters from previous positive results, increasing the number of points involved and ensuring most points lie within a discrete subgroup. The constructed functions are smooth and rapidly decaying, but not analytic or super-exponentially decaying, which aligns with existing positive results.

Role of AI in the Discovery

The resolution of the HRT conjecture was assisted by artificial intelligence. AI was used to develop the initial proof strategy, with the final arguments and overview written by human researchers. Additionally, traditional numerical computations were employed to verify specific steps of the argument, complementing the AI-driven approach. The authors disclosed their use of AI, providing details on methods and relation to past literature.

Implications for Mathematical Research

This development marks a significant step in the field of time-frequency analysis and demonstrates the increasing utility of AI as a tool in complex mathematical problem-solving. The responsible disclosure and integration of AI methods, alongside traditional verification, set a precedent for future research in mathematics.