View Single Post
  #4  
Old 04-16-2012, 01:47 PM
htlin's Avatar
htlin htlin is offline
NTU
 
Join Date: Aug 2009
Location: Taipei, Taiwan
Posts: 601
Default Re: what is bias means?

Quote:
Originally Posted by antinucleon View Post
Thanks dear prof first!
Assume we start at init hypothesis w=(0, 0, 0), init target function is w_t=(0.5, 2, 5)
As w = w + yx, y equals 0 or 1,x_n = (1, 0.5, 3) etc
When in iteration, w_0 = w_0 + y*x_0, as y and x_0 can only be 0 or 1, so the hypothesis function's threshold may be quite different from the target function, right?
In your case, the target function is equivalently (1, 4, 10), (2, 8, 20), so it can still be implemented with the PLA model even with integer w_0. But of course, learning is not for replicating the target function, but for approximating it. In practice the learned hypothesis is somewhat different from the target function more often than not. Hope this helps.
__________________
When one teaches, two learn.
Reply With Quote