Revised version of the sidelength 3 case:
Matrices:
If we start to look at the determinant... Also, just as a note, this is how I was numbering the cells [in increasing order from left to right by row.]
Here is the side length of 2 case:
The determinant is clearly 0. One thing we can note about the starting matrix, is that it has a diagonal of 0s and it is symmetrical over that diagonal. Since we can take either row or column to do the determinant by, in the following case, there is always either a row or column that has only one 1 in it. If we take that, it is simple to do the determinant.
Here is the side length 3 case:
Now, we see that there is a column of all 0s. If we did the determinant by this column, the result will clearly be 0.
To make the result more clear: (The green is the column I went down to take the determinant, the orange is the row that was ignored for the only 1 there.)
...not sure how helpful this is.
Thinking about Triangles:
In any given triangle, we know that the bottom row looks like this:
Now, if we have the numbers filled in like this:
So for the kth cell in this row, the number a_k is equal to 1 if k is even or 0 if k is odd.
If k is odd, then the kth slot is a 0 and the (k-1)th and (k+1)th slots are 1. The rest of the grid is filled with Os, so all Xs have to be in the bottom row. From the 1s, we know there has to be an X in some of the following slots: (k-2), k, (k+2). This particular area is ambiguous.
If k is even, then the kth slot is a 1 and the (k-1)th and (k+1)th are 0. From here, we know that either (k-1) or (k+1) contains an X.






I love the pictures. Some of them could definitely be used on the poster.
ReplyDeleteThis relates to: (Triangles in a triangle arrangement with the neighborhood defined by edges and not counting self.) I think this should be specified, as a week later, I had to go back to check what case was being done.
I don't see anything else to add. This should be one section or subsection of the poster.