Cloudera
DS-200 · Question #17
When optimizing a function using stochastic gradient descent, how frequently should you update your estimate of the gradient?
The correct answer is A. Once after every pass through the data set C. For each observation with a probability that you choose ahead of time. See the full explanation below for the reasoning.
Question
When optimizing a function using stochastic gradient descent, how frequently should you update your estimate of the gradient?
Options
- AOnce after every pass through the data set
- BOnce per observation
- CFor each observation with a probability that you choose ahead of time
- DAfter a random number of observations
- EOnce every N observations, where you decide N ahead of time
How the community answered
(29 responses)- A79% (23)
- B3% (1)
- D10% (3)
- E7% (2)
Community Discussion
No community discussion yet for this question.