Perceptrons: Functions not learnable without bias
-
05-11-2019 - |
Pergunta
I am trying to determine which functions are not learnable without a bias when building a perceptron. The set of functions I need to evaluate is {NOT, OR, AND}.
Could someone help interpret these results?
AND FUNCTION
OR FUNCTION
I get a divide by 0 error in my script.
NOT FUNCTION
It would appear that the AND function does not crash like the OR function but the end result is incorrect. The OR function simply won't graph due to a divide by 0 error. The NOT function appears to still be learnable.
If desired, I can post my python script with how I generated these graphs. I am using a sigmoid activate function.
Nenhuma solução correta
Licenciado em: CC-BY-SA com atribuição
Não afiliado a cs.stackexchange