I have been using the weight update function from Russell and Norvig's AI book:
Wj <- Wj + alpha x Err x g'(in) x Xj
Where g' is the derivative of the activation function (I think it can be ignored?), and alpha is just for scaling the amount of change. Xj is the j'th input bit. Err is the...
In order to avoid looking at every point, the points need to be organized in some way. Here's how I would do it -
Make a bounding box, and break it into ( 40? 200?) equal squares. Each square stores pionters to the points currently in it. To look for the closest unit, first check for other...
This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
By continuing to use this site, you are consenting to our use of cookies.