Advanced Learning Algorithm 12: Back Propagation

brandon·2023년 8월 24일
0

SupervisedML

목록 보기
21/27

1. Computation Graph

  • Back propagation is basically a process to find a relationship between cost J function and the parameters.
  • It works its way backwards from the result, then finds the derivatives values as it moves backwards.
  • N (Number of units) + P (Number of parameters) = the number of steps needed to compute the derivatives.
    What derivatives??
profile
everything happens for a reason

0개의 댓글