1 | initial version |
For each nonnegative integer $k$, you could find the possible tuples of matrices $(A_1, ..., A_k)$ as follows:
for each of -1
, 0
, 1
, find the ways to write it
as a sum of $k$ summands each in {-1
, 0
, 1
}.
the target matrix $M$ has $n^2$ entries $m_{i, j}$; so the $(A_1, ..., A_k)$ are obtained by combining all the possible ways to get each entry.