4.3 Minimising errors (difficult)
Minimising the error in the more complex neural network on this page may seem impossible at first glance. But in fact, at least for many Boolean functions it is not as difficult as you might think. If you haven't yet developed an intuition for how to “construct paths” in order to succeed, you can continue practising with the simpler networks.
Instructions
- Use the checkboxes to select which training data should be loaded.
- Click
Newto select the preset values for a simple separation, a new randomly selected Boolean function or new random numbers as a data set. - Now click on the plus or minus signs in the figure to change the weights or the threshold value. The current error of the network is automatically displayed. If the error is small enough, you will receive a pop-up message with your required time.
- Click
Resetto undo all your current changes to the weights and threshold. - Click
Trainto start or resume training. - In the figure on the left, blue stands for negative values and red stands for positive values.
Tasks
- First try to master the pre-selected data set.
- Then, experiment with several different examples from the Boolean functions data set and minimise the network’s error for each one.
- Can you also minimise error using a data set of random numbers?
Share this page
