This Machine Learning MCQ questions and answers assess your understanding of algorithms, models, and real-world ML applications.

Machine Learning MCQ

Which of the following best indicates underfitting?

Improves model evaluation by averaging over folds

Why is cross-validation used in model evaluation?

Model has low training error and high testing accuracy

Model performs well on training data but poorly on new data

What does overfitting in machine learning mean?

Training data builds model, testing data evaluates it

What is the difference between training and testing data?

Tradeoff between training and testing speed

It’s a tradeoff between underfitting and overfitting

What does the bias-variance tradeoff refer to?

Reduces overfitting and improves generalization

Enables systems to learn from data without being explicitly programmed

To evaluate model performance on unseen data

Why do we split data into training and testing sets?

High training accuracy but low test accuracy

Penalizes complex models to prevent overfitting

What is the purpose of regularization in machine learning?

Which of the following are major types of Machine Learning?

Training data is used to learn, testing data to evaluate

Testing data is used to tune hyperparameters

Training data contains only numeric features

How does training data differ from testing data?

One complete pass through the training data

What is an epoch in machine learning training?

What is the purpose of a loss function in machine learning?

Model performs poorly on both training and test data

Model performs well on training data but poorly on unseen data

Reducing both training and testing errors to zero

Using only complex models to avoid variance

What is the bias-variance tradeoff in machine learning?

A model performing well on training data but poorly on testing data

Which scenario best describes underfitting?

What does the 'label' represent in supervised learning?

Supervised learning does not require training

Supervised learning uses labeled data, unsupervised does not

Unsupervised models always perform better

What is the key difference between supervised and unsupervised learning?