6.034 Artificial Intelligence - Recitations, fall 2004 online slides on learning

Next: Examples Previous: The AND Function

Function Learned By One Layer Net

The final network can be expressed as an equation of the parameters x1 through xn

w1 x1 + w2 x2 + ... + wn xn = threshold

If learned, the network can be represented as a hyperplane

             + = Class 1   +     |   +/w1x1 + ... + wnxn = t
             - = Class 2+        |++ /
                        +  +     |+ /
                                 | /   -  -
                       +  +   +  |/  -    -  -
                     ------------/------------
                        +   +   /|     -
                         +     /-| -   -  -
                      + + + + /  |  -    -
                          +  /  -|
                                 |