MCQ Collection
India MCQs
Practice India questions with answers and explanations.
Choose an option to check your answer.
A.
Euclidean area
B.
Hinge length only
C.
Binary cross-entropy
D.
Mean absolute temperature
Show Answer
Correct Answer: C. Binary cross-entropy
Explanation:
Binary cross-entropy matches Bernoulli likelihood.
Choose an option to check your answer.
A.
Unseen data
B.
Constant arrays only
C.
No data
D.
Its own labels only
Show Answer
Correct Answer: A. Unseen data
Explanation:
Overfitting harms generalisation.
Choose an option to check your answer.
A.
Euclidean area
B.
Binary cross-entropy
C.
Mean absolute temperature
D.
Hinge length only
Show Answer
Correct Answer: B. Binary cross-entropy
Explanation:
Binary cross-entropy matches Bernoulli likelihood.
Choose an option to check your answer.
A.
Its own labels only
B.
Unseen data
C.
Constant arrays only
D.
No data
Show Answer
Correct Answer: B. Unseen data
Explanation:
Overfitting harms generalisation.
A. Precision is TP/(TP+FP).B. Recall is TP/(TP+FN).C. Accuracy is always suitable for highly imbalanced data.D. Specificity is TN/(TN+FP).
Show Answer
Correct Answer: A|B|D
Explanation:
Accuracy can be misleading under severe imbalance.
A. The learning rate controls step size.B. A very large learning rate can cause divergence.C. Every non-convex objective has one unique global minimum.D. Feature scaling can improve convergence.
Show Answer
Correct Answer: A|B|D
Explanation:
Non-convex objectives may have many local features.
A. AB is m×p.B. BA is always defined.C. Matrix multiplication is generally not commutative.D. Computing AB by the standard algorithm uses O(mnp) scalar operations.
Show Answer
Correct Answer: A|C|D
Explanation:
BA requires p=m.
A. Precision is TP/(TP+FP).B. Recall is TP/(TP+FN).C. Accuracy is always suitable for highly imbalanced data.D. Specificity is TN/(TN+FP).
Show Answer
Correct Answer: A|B|D
Explanation:
Accuracy can be misleading under severe imbalance.
A. The learning rate controls step size.B. A very large learning rate can cause divergence.C. Every non-convex objective has one unique global minimum.D. Feature scaling can improve convergence.
Show Answer
Correct Answer: A|B|D
Explanation:
Non-convex objectives may have many local features.
Choose an option to check your answer.
A.
Its own labels only
B.
Constant arrays only
C.
No data
D.
Unseen data
Show Answer
Correct Answer: D. Unseen data
Explanation:
Overfitting harms generalisation.
Choose an option to check your answer.
A.
Euclidean area
B.
Binary cross-entropy
C.
Hinge length only
D.
Mean absolute temperature
Show Answer
Correct Answer: B. Binary cross-entropy
Explanation:
Binary cross-entropy matches Bernoulli likelihood.
Choose an option to check your answer.
A.
Hinge length only
B.
Binary cross-entropy
C.
Mean absolute temperature
D.
Euclidean area
Show Answer
Correct Answer: B. Binary cross-entropy
Explanation:
Binary cross-entropy matches Bernoulli likelihood.
