On the complexity of integer matrix multiplication

On the complexity of integer matrix multiplication
复制标题

关于整数矩阵乘法的复杂度

DOI:
10.1016/j.jsc.2017.11.001
复制
发表时间:
2017
期刊:
J. Symb. Comput.
影响因子:
--
通讯作者:
J. Hoeven
J. Hoeven
中科院分区:
--
文献类型:
--
作者:
David Harvey;J. Hoeven

文献摘要

被引文献

相似文献

令 M (n) 表示 n 位整数相乘的位复杂度,让 ωε(2, 3] 为矩阵乘法的指数,让 lg⁎⁡ n 为迭代对数。假设 log⁡ d= O (n) 并且 M (n)/(n log⁡ n) 递增,我们证明具有 n 位整数项的 d× d 矩阵可以在 O 中相乘(d 2 M (n)+ d ω n 2 O (lg⁎⁡ n− lg⁎⁡ d) M (lg⁡ d)/lg⁡ d) 位运算特别是,如果 n 比 d 大,比如 d= O (log⁡ n),则复杂度仅为 O (d 2 M (n))。
Let M (n) denote the bit complexity of multiplying n-bit integers, let ω∈(2, 3] be an exponent for matrix multiplication, and let lg⁎⁡ n be the iterated logarithm. Assuming that log⁡ d= O (n) and that M (n)/(n log⁡ n) is increasing, we prove that d× d matrices with n-bit integer entries may be multiplied in O (d 2 M (n)+ d ω n 2 O (lg⁎⁡ n− lg⁎⁡ d) M (lg⁡ d)/lg⁡ d) bit operations. In particular, if n is large compared to d, say d= O (log⁡ n), then the complexity is only O (d 2 M (n)).