arXiv:2611.14743v1 Announce Type: new Abstract: We present schubmult, a GPL-licensed Python system for exact multiplication of ordinary, double, quantum, and quantum double Schubert polynomials, as well as independent-alphabet triple and quantum triple products. Its central algorithm expands one factor into filtered column paths and couples them to Pieri computations on the other factor. Dynamic programming shares intermediate states, and the algorithm returns every Schubert-basis coefficient as an exact arithmetic circuit. We prove input-sensitive operation bounds for the entire product, not just a selected coefficient: for inputs of minimal rank $n$, the classical algorithms have complexity $2^{O(n\log n)}$ and polynomial complexity in $n$ for fixed Lehmer-code lengths. The quantum algorithms have the same general bound and polynomial complexity when, in addition, the number of nonzero entries in the Lehmer code of $v^{-1}$ is fixed. We describe the implementation and distinguish it from the algorithms analyzed, verify computations against independent routines and exact identities, and report reproducible timings for flag and Grassmannian workloads. Parabolic quantum products are treated separately via the Peterson-Woodward comparison.
Continue to arXiv Mathematics →You are being forwarded in 6s — stay on this page