Machine Learning Quiz

Which of the following are supervised learning problems? (Multiple Correct)

• Clustering Spotify users based on their listening history

• Weather forecast using data collected by a satellite

• Predicting tuberculosis using patient's chest X-Ray

• Training a humanoid to walk using a reward system

Which of the following are regression tasks? (Multiple Correct)

• Predicting the outcome of an election

• Predicting the weight of a giraffe based on its height

• Predicting the emotion conveyed by a sentence

• Identifying abnormal data points

Which of the following are classification tasks? (Multiple Correct)

• Predicting the outcome of an election

• Predicting the weight of a giraffe based on its height

• Predicting the emotion conveyed by a sentence

• Identifying abnormal data points

Which of the two functions overfit the training data?

• Both functions F1 & F2

• Function F1

• Function F2

• None of them

Which of the two functions will yield higher training error?

• Function F1

• Function F2

• Both functions F1 & F2 will have the same training error

• Cannot be determined

What does the term 'policy' refer to in reinforcement learning?

• A set of rules governing the environment

• The reward function

• The initial state of the environment

• The strategy the agent follows to choose actions

Given the following dataset, for k = 3, use KNN regression to find the prediction for a new data-point (2, 3). (Use Euclidean distance measure for finding closest points.)

• 2.0

• 2.6

• 2.8

• 3.2

For any given dataset, comment on the bias of k-nearest classifiers upon increasing the value of k.

• The bias of the classifier decreases

• The bias of the classifier does not change

• The bias of the classifier increases

• Cannot be determined

Bias and variance are given by:

• E[\hat{f}(x)] - f(x), E[(E[\hat{f}(x)] - \hat{f}(x))^2]

• E[\hat{f}(x)] - f(x), E[(E[\hat{f}(x)] - \hat{f}(x))]^2

• (E[\hat{f}(x)] - f(x))^2, E[(E[\hat{f}(x)] - \hat{f}(x))^2]

• (E[\hat{f}(x)] - f(x))^2, E[(E[\hat{f}(x)] - \hat{f}(x))]^2

Which of the following statements are FALSE regarding bias and variance? (Multiple Correct)

• Models that overfit have high bias

• Models that overfit have low bias

• Models that underfit have high variance

• Models that underfit have low variance