Neural Networks w/ JAVA - Tutorial 04

00:06 have 3 inputs + a bias and need to obtain equation of a plane separating the 0s and 1s

00:35 step #0 randomly initialize weights w0, w1, w2, and w3

00:45 step #1 calculate weighted sum

01:00 step #2 apply activation function

01:18 step #3 determine error

01:54 'learning rate' is the rate at which the neural network learns (ranges from 0 to 1)

01:27 step #4 adjust weights

02:31 objective here is to determine what weights would

lead to 'Target Result' = 'Result' for all vectors in training data

02:02 repeat steps 1 to 4 until error = 0

03:32 demo prebuilt version of the app.

05:09 code the application

10:46 test run completed application

-------------------------------------------------------------------------------------------------------------

quickly download, import into Eclipse, and run zip file for

'Neural Networks w/ JAVA - Tutorial 04'

-------------------------------------------------------------------------------------------------------------

-----------------------------------------------------------------------------------------------------------

download source code @ https://sites.fastspring.com/prototypeprj/instant/2020 download source code @ https://sites.fastspring.com/prototypeprj/instant/all

-----------------------------------------------------------------------------------------------------------