Practice free →
HomeClaudeaifoundationslinear_regression_and_supervised_learning › The resulting per-example log-likelihood is, up …

The resulting per-example log-likelihood is, up to constants

AProportional to −(y⁽ⁱ⁾ − θᵀx⁽ⁱ⁾)²
BProportional to |y⁽ⁱ⁾|
CIndependent of θ
DProportional to log(θᵀx⁽ⁱ⁾)
Answer & Solution
Correct answer: A. Proportional to −(y⁽ⁱ⁾ − θᵀx⁽ⁱ⁾)²
log p(y|x; θ, σ) under Gaussian noise = -1/(2σ²)·(y - θᵀx)² + const. Maximising the log-likelihood w.r.t. θ is equivalent to minimising the squared-error sum — derives least-squares as MLE.
Solve this in the app — Claude practice & 24k+ MCQs →
Related questions