Hi, I am finding this book a great read. I understand the author posts here, so I have a couple of questions about the text that have thrown me a little. 1. In chapter 7, page 256 the first paragraph reads... ''Don''t forget that the last weight in each neuron''s weight vector is the weight for the bias, which as we have already discussed, is always set to -1''. This is confusing to me, as the previous text explains why it should be ''multiplied'' by -1, but the bias for the neuron is an individual value. Seems there may be some confusion here between the bias for the neuron, and the bias for the bias (which is for the network as a whole). 2. Further on the same page, encoding the network reads... ''I did not include a bias in this network just to keep things simple''. ''A'' bias? Should this more read ''I did not include the bias for each neuron in this network...''. If that is the case, should the bias for the neuron be multiplied by -1 also before it is store in the genome? Many thanks for any clarification. Rael

Hi Raeldor,

I''m glad you are enjoying my book but this forum is the wrong place to ask specific questions about the text.

The ai-junkie site has a forum so I will transfer this question there before answering it. This is the URL:

http://pub59.ezboard.com/fstimulatefrm1

see you there...





My Website: ai-junkie.com | My Book: AI Techniques for Game Programming