nerdexam
IBM

C1000-012 · Question #52

How would someone train a neural network?

The correct answer is D. Back propagate the network output through the network and forward propagate errors in order to update relevant weightings. See the full explanation below for the reasoning.

Question

How would someone train a neural network?

Options

  • ACompare results from forward propagation to an error score and update weightings accordingly using back propagation
  • BCompare input of network to output of network and back propagate the difference to the input connections layer
  • CCalculate the error rate using cognitive dissonance and back propagate errors to the neurons to minimize the error rate
  • DBack propagate the network output through the network and forward propagate errors in order to update relevant weightings

How the community answered

(30 responses)
  • A
    10% (3)
  • B
    3% (1)
  • C
    3% (1)
  • D
    83% (25)

Community Discussion

No community discussion yet for this question.

Full C1000-012 Practice