Claude linear_regression_and_supervised_learning — practice questions
15 free MCQs with worked solutions. Tap any question for the answer + explanation, or practice them all in the app.
Practice Claude linear_regression_and_supervised_learning in the app →The SUPERVISED-learning setup gives the learnerLinear regression assumes the hypothesis function h_θ(x) takes the formThe least-squares cost function in CS229 linear-regression notes isThe gradient-descent update rule for parameter θⱼ in linear regression is'Α' in θⱼ := θⱼ − α·∂J/∂θⱼ is called theBATCH gradient descent differs from STOCHASTIC gradient descent in thatThe LMS (Widrow–Hoff) update rule on a single training example isThe closed-form 'normal equations' solution for least-squares linear regression isThe design matrix X for n training examples with d features (plus intercept) has shapeWhy is the least-squares optimisation for LINEAR regression guaranteed to converge to the GLOBAL minimum underThe PROBABILISTIC justification for the least-squares cost assumes the noise ε⁽ⁱ⁾ in y⁽ⁱ⁾ = θᵀx⁽ⁱ⁾ + ε⁽ⁱ⁾ isLogistic regression uses which hypothesis function?Closed-form normal equations require XᵀX to beWhy is STOCHASTIC gradient descent often preferred over BATCH gradient descent on very large datasets?The resulting per-example log-likelihood is, up to constants