The parameters
The network parameters, to be optimized/learned:
- \boldsymbol{a} represents the visible bias, a vector of same length M as \boldsymbol{x} .
- \boldsymbol{b} represents the hidden bias, a vector of same length N as \boldsymbol{h} .
- \boldsymbol{W} represents the interaction weights, a matrix of size M\times N .
Note that we have specified the lengths of bm{x} and \boldsymbol{h} . These
lengths define the number of visible and hidden units, respectively.