🧠

Artificial Intelligence & Machine Learning MCQ

Test your AI & Machine Learning knowledge with 100 multiple choice questions covering fundamentals to advanced concepts, with instant feedback and explanations.

100 Questions 40 Beginner 40 Intermediate 20 Advanced
1

What is the transformer architecture and why did it revolutionize AI?

2

What is BERT and how does it differ from GPT?

3

What is fine-tuning a pre-trained model?

4

What is a GAN (Generative Adversarial Network)?

5

What is a VAE (Variational Autoencoder)?

6

What is word2vec?

7

What is the vanishing gradient problem?

8

What is LSTM and how does it address the vanishing gradient problem?

9

What is object detection and how does it differ from image classification?

10

What is the purpose of residual connections (skip connections) in deep networks?

11

What is RLHF (Reinforcement Learning from Human Feedback)?

12

What is prompt engineering?

13

What is RAG (Retrieval-Augmented Generation)?

14

What is the difference between precision-recall and ROC curves?

15

What is gradient clipping?

16

What is knowledge distillation?

17

What is neural architecture search (NAS)?

18

What is a recommendation system?

19

What is concept drift in machine learning?

20

What is federated learning?

21

What is a vector database and why is it used in AI applications?

22

What is LoRA (Low-Rank Adaptation)?

23

What is the difference between model accuracy and fairness in ML?

24

What is mean squared error (MSE) vs mean absolute error (MAE)?

25

What is model explainability (XAI)?

26

When should you choose a decision tree over logistic regression for a classification task?

27

What is the elbow method used for in clustering?

28

What is stratified sampling and why is it used when splitting a dataset?

29

Why is one-hot encoding commonly used for categorical features in machine learning?

30

What is the main practical tradeoff when increasing the depth of a decision tree?

31

In scikit-learn-style workflows, what is the purpose of a Pipeline object?

32

What is early stopping during neural network training?

33

What is the difference between bagging and boosting as ensemble techniques?

34

Why might you use the Adam optimizer instead of plain stochastic gradient descent (SGD) when training a neural network?

35

What is data leakage in a machine learning workflow?

36

What is the purpose of the learning rate schedule (e.g., step decay or cosine annealing) during training?

37

What is tokenization in the context of NLP and language models?

38

What is the purpose of a learning curve (training size vs. error) when diagnosing a model?

39

Why is min-max scaling or standardization often applied before training algorithms like k-NN or SVM?

40

What is the practical difference between using grid search and random search for hyperparameter tuning?