Machine Learning Quiz

Which of the following best differentiates supervised from unsupervised learning?

• Supervised learning uses labeled data; unsupervised learning uses unlabeled data.

• Supervised learning uses unlabeled data; unsupervised learning uses labeled data.

• Both methods use only labeled data.

• Both methods use only unlabeled data.

Which of the following activation functions is used in neural networks?

• ReLU

• Sigmoid

• Softmax

• All of the above

A higher learning rate always improves model accuracy.

• True

• False

• It depends on the dataset

• Cannot be determined

What is the primary purpose of gradient descent in machine learning?

• To maximize the cost function.

• To minimize the cost function by iteratively updating parameters.

• To compute the derivative of the cost function without any updates.

• To randomly initialize model weights.

Which of the following best describes overfitting in a machine learning model?

• High performance on training data and poor performance on unseen data.

• High performance on both training and testing data.

• Poor performance on both training and testing data.

• High performance on testing data and low performance on training data.

Which algorithm is primarily used for classification tasks among the following?

• K-Means Clustering

• Random Forest Classifier

• Principal Component Analysis

• Linear Regression

What distinguishes batch gradient descent from stochastic gradient descent?

• Batch uses a subset of data per iteration, stochastic uses the entire dataset.

• Batch uses the entire dataset per iteration, stochastic uses one sample per iteration.

• Batch is faster than stochastic.

• There is no difference.

Which evaluation metric is best suited for imbalanced classification datasets?

• Accuracy

• Mean Squared Error

• Precision-Recall

• R-squared

What is the primary advantage of using Convolutional Neural Networks (CNNs) for image processing?

• Ability to capture spatial hierarchies in images.

• Reduced number of parameters due to shared weights.

• Inherent translation invariance.

• All of the above

What is the main purpose of using dropout in neural networks?

• To speed up training by reducing the number of neurons.

• To prevent overfitting by randomly deactivating neurons during training.

• To improve model accuracy by increasing model complexity.

• To enhance data augmentation.