The loss function of logistic regression is the cross-entropy loss function.

Question

Options

• ATRUE
• BFALSE

• A
  73% (32)
• B
  27% (12)

Explanation

Logistic regression uses the cross-entropy loss (also called log loss) because the model outputs probabilities via the sigmoid function, and cross-entropy measures the divergence between predicted probabilities and true binary labels - making it the natural fit derived from maximum likelihood estimation. Minimizing cross-entropy is mathematically equivalent to maximizing the likelihood of the observed data under a Bernoulli distribution, which is exactly what logistic regression models. Option B is wrong because there is no alternative loss function in standard logistic regression; mean squared error (MSE) is sometimes confused here, but MSE is non-convex when applied to sigmoid outputs and leads to poor optimization. Memory tip: Associate "logistic" with "log" - cross-entropy loss contains a logarithm (-y·log(p) - (1-y)·log(1-p)), so if you see logistic regression, think log loss = cross-entropy.

No community discussion yet for this question.