A complexity chasm for solving univariate sparse polynomial equations over $p$-adic fields
Abstract
We reveal a complexity chasm, separating the trinomial and tetranomial cases, for solving univariate sparse polynomial equations over certain local fields. First, for any fixed field , we prove that any polynomial with exactly monomial terms, degree , and all coefficients having absolute value at most , can be solved over in deterministic time in the classical Turing model. (The best previous algorithms were of complexity exponential in , even for just counting roots in .) In particular, our algorithm generates approximations in with bit-length to all the roots of in , and these approximations converge quadratically under Newton iteration. On the other hand, we give a unified family of tetranomials requiring digits to distinguish the base- expansions of their roots in .
Cite
@article{arxiv.2003.00314,
title = {A complexity chasm for solving univariate sparse polynomial equations over $p$-adic fields},
author = {J. Maurice Rojas and Yuyu Zhu},
journal= {arXiv preprint arXiv:2003.00314},
year = {2021}
}
Comments
19 pages, 3 figures. This version contains an Appendix missing from the ISSAC 2021 conference version, as well as some corrections and improvements