A near-optimal algorithm for computing real roots of sparse polynomials
M Sagraloff - Proceedings of the 39th International Symposium on …, 2014 - dl.acm.org
Proceedings of the 39th International Symposium on Symbolic and Algebraic …, 2014•dl.acm.org
Let p∈ Z [x] be an arbitrary polynomial of degree n with k non-zero integer coefficients of
absolute value less than 2τ. In this paper, we answer the open question whether the real
roots of p can be computed with a number of arithmetic operations over the rational numbers
that is polynomial in the input size of the sparse representation of p. More precisely, we give
a deterministic, complete, and certified algorithm that determines isolating intervals for all
real roots of p with O (k 3⊙ log (n τ)⊙ log n) many exact arithmetic operations over the …
absolute value less than 2τ. In this paper, we answer the open question whether the real
roots of p can be computed with a number of arithmetic operations over the rational numbers
that is polynomial in the input size of the sparse representation of p. More precisely, we give
a deterministic, complete, and certified algorithm that determines isolating intervals for all
real roots of p with O (k 3⊙ log (n τ)⊙ log n) many exact arithmetic operations over the …
Let p ∈ Z[x] be an arbitrary polynomial of degree n with k non-zero integer coefficients of absolute value less than 2τ. In this paper, we answer the open question whether the real roots of p can be computed with a number of arithmetic operations over the rational numbers that is polynomial in the input size of the sparse representation of p. More precisely, we give a deterministic, complete, and certified algorithm that determines isolating intervals for all real roots of p with O(k3·log(nτ)·logn) many exact arithmetic operations over the rational numbers.
When using approximate but certified arithmetic, the bit complexity of our algorithm is bounded by Õ(k4·n·(τ+k)), where Õ(·) indicates the omission of logarithmic factors. Hence, for sufficiently sparse polynomials (i.e. k = O(logc(nτ)) for a constant c), the bit complexity is Õ(nτ), which is optimal up to logarithmic factors.
ACM Digital LibraryShowing the best result for this search. See all results