- Ok, this is a total beginner question. Neural Networks are designed by specifying the input, and the output, and it eventually builds the network saying "for this specific input and output, here are ways to achieve that result" and after a while of inputting that data (both input and output values), patterns immerge in the calculations which allow the network to be very accurate? Correct? (I hope I explained that ok)
Assuming that, and assuming I would want to program that in C++, how would I do that? Am I correct in thinking that I would have to build all possible calculations for any given output? If so, whats the point? That leaves a great deal of possibilities open to things I havent thought of. I guess my real qustion is how would a NN be written in C++ (like a really quick example would help)?
[Non-text portions of this message have been removed]
- For coding, try: http://google.com/codesearch?q=neural+net+lang%3Ac%2B%2B
And for beginners information, try: