迭代矩阵乘法的齐次深度五下界

无需登录。完整内容与个人兴趣标记的公开阅读入口,可直接交给 GPT 阅读。

公开快照同步: · 内容整理:2026-10-19T08:04:09+08:00

OpenAI论文 / 理论计算机科学 · 2026-09-25

迭代矩阵乘法的齐次深度五下界

Homogeneous depth-five lower bounds for iterated matrix multiplication

OpenAI 数学仓库收录的研究手稿。展开查看英文摘要,或标记后让我进一步解释。

具体内容

该稿件研究迭代矩阵乘法IMM_{n,n}在齐次深度五算术电路中的门复杂度。作者声称:在任意特征零域上,每个语法齐次ΣΠΣΠΣ电路计算IMM_{n,n}至少需要n^{√n/400}个门(n充分大,阈值与域无关);底层线性形式可任意支撑,允许任意有限扇入、扇出与门共享。摘要还称在任意域上块展开给出至多n^{√n+4}个门的电路,故特征零域上门复杂度为n^{Θ(√n)}。意义在于给出接近紧的深度五下界,但尚未经独立审查。

AI 产生的研究手稿;收录不表示已完成独立审查或同行评议。

英文原始摘要

Let $`\mathop{\mathrm{IMM}}\nolimits _{n,n}`$ be the $`(1,1)`$ entry of a product of n independent $`n\times n`$ matrices of variables. Over every field of characteristic zero, every syntactically homogeneous $`\Sigma\Pi\Sigma\Pi\Sigma`$ circuit computing $`\mathop{\mathrm{IMM}}\nolimits _{n,n}`$ has at least $`n^{\sqrt n/400}`$ gates for all sufficiently large n, with an absolute threshold independent of the field. Bottom linear forms may have arbitrary support, and arbitrary finite fan-in, fan-out, and gate sharing are allowed. Over every field, a block expansion gives such circuits with at most $`n^{\sqrt n+4}`$ gates for n ≥ 2. Thus the gate complexity over characteristic-zero fields is $`n^{\Theta(\sqrt n)}`$.

用户标记:未标记

固定编号:1c5734748640e1d2

原文来源