What is the best way to determine the learning rate parameters for stochastic gradient descent when the distribution of the input data shifts over time?

Question

Options

• AThe learning rate should be adjusted periodically based on the setting that optimizes the objective
• BThe learning rate should be fixed number that decays as the number of observations in the data
• CThe learning rate should be the value that optimizes the value of the objective function over the
• DThe learning rate should be a fixed number with a constant decay factor
• EThe learning rate should be continuously adjusted based on the value that optimizes the objective

• A
  3% (1)
• B
  6% (2)
• C
  78% (28)
• D
  11% (4)
• E
  3% (1)

No community discussion yet for this question.