Let denote the matrix multiplication tensor (and write ) and let denote the determinant polynomial considered as a tensor. For a tensor , let denote its border rank. We (i) give the first hand-checkable algebraic proof that $\underline R(M_2)=7\underline R(M_{\langle 223\rangle})=10\underline R(M_{\langle 233\rangle})=14$, where previously the only nontrivial matrix multiplication tensor whose border rank had been determined was ,(iii) prove , (iv) prove , improving the previous lower bound of , (v) prove $\underline R(M_{\langle 2nn\rangle})\geq n^2+1.32nn\geq 25\underline R(M_{\langle 2nn\rangle})\geq n^2+14\leq n\leq 25\underline R(M_{\langle 3nn\rangle})\geq n^2+2 n+1 n\geq 21\underline R(M_{\langle 3nn\rangle})\geq n^2+24\leq n\leq 21$. Our results utilize a new technique initiated by Buczy'{n}ska and Buczy'{n}ski, called border apolarity. The two key ingredients are: (i) the use of a multi-graded ideal associated to a border rank decomposition of any tensor, and (ii) the exploitation of the large symmetry group of to restrict to -invariant ideals, where is a maximal solvable subgroup of the symmetry group of .

生成讲解博客的中文版本

生成AI智能校验的中文版本