Home › Claude › aifoundations › linear_regression_and_supervised_learning › BATCH gradient descent differs from STOCHASTIC g…
BATCH gradient descent differs from STOCHASTIC gradient descent in that
ABatch only works for classification; stochastic only for regression
BBatch uses one example per update; stochastic uses the entire training set per update
CBatch uses the entire training set per update; stochastic uses one example per update
DBatch is non-iterative; stochastic is iterative
Answer & Solution
Correct answer: C. Batch uses the entire training set per update; stochastic uses one example per update
Batch (a.k.a. full-batch) sums over all n examples each step. Stochastic (SGD) updates after each single example — faster early progress but oscillates around the minimum unless α decays.
Related questions
The resulting per-example log-likelihood is, up to constantsWhy is STOCHASTIC gradient descent often preferred over BATCH gradient descent on very larClosed-form normal equations require XᵀX to beLogistic regression uses which hypothesis function?The PROBABILISTIC justification for the least-squares cost assumes the noise ε⁽ⁱ⁾ in y⁽ⁱ⁾ Why is the least-squares optimisation for LINEAR regression guaranteed to converge to the The design matrix X for n training examples with d features (plus intercept) has shapeThe closed-form 'normal equations' solution for least-squares linear regression is