I don't get this matrix calculation from an ANN book

Recommended Posts

Hi im reading "C++ Neural Networks and Fuzzy Logic by Valluru B. Rao", anyway in one section, i don't get how they are getting this answer for the product of the matrix; ...from the book
Quote:
 First, we notice that the binary to bipolar mapping of A = (1, 0, 1, 0) gives the vector (1, –1, 1, –1). Then we take the transpose, and multiply, the way matrices are multiplied, and we see the following:
     1  [1   -1   1   -1]       1   -1   1   -1
1                     =   -1    1  -1    1
1                          1   -1   1   -1
1                         -1    1  -1    1


to me it looks like the answer should be: 1 -1 1 -1 1 -1 1 -1 1 -1 1 -1 1 -1 1 -1 I don't get it. Whats going on? Thanks

Share on other sites
The result is OK, but there is a mistake in the transpose (probably a typo). It should be
     1  [1   -1   1   -1]      1   -1   1   -1    -1                     =   -1    1  -1    1     1                          1   -1   1   -1    -1                         -1    1  -1    1

Hope it helps

ohhh, thanks

Create an account

Register a new account

• Forum Statistics

• Total Topics
628391
• Total Posts
2982419

• 10
• 9
• 19
• 24
• 11