My understanding is that the conjugate gradient method is faster than gradient descent because it does less zig zags while descending. How come the state of the art papers I see all use gradient descent for backpropagation and not conjugate gradient descent?

  • $\begingroup$ Isn't conjugate gradient a descent strategy. I'm not sure what two things you are comparing, and I'm not sure what propagation strategies have to do with that comparison. Can you add details to clearly identify the two things being compared? $\endgroup$ – FelicityC Nov 7 '18 at 18:38
  • $\begingroup$ @FelicityC I meant conjugate gradient descent. Thanks for asking. I have edited the question to clarify this. $\endgroup$ – FullStack Nov 7 '18 at 22:16

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.