Podcast
Questions and Answers
Which of the following tasks are examples of unsupervised learning?
Which of the following tasks are examples of unsupervised learning?
- Predicting the likelihood of a customer clicking 'like' on a product, given their demographic information.
- Identifying social clusters within a community. (correct)
- Sorting news articles into predefined categories based on their titles.
- Predicting the exchange rate between USD and EURO based on historical data.
Which of the following is TRUE regarding Reinforcement Learning (RL)?
Which of the following is TRUE regarding Reinforcement Learning (RL)?
- The primary goal of an RL policy is immediate reward, ignoring long-term consequences.
- RL excels in creating self-improving game-playing agents, as shown in strategic games. (correct)
- During training, an RL agent needs an external supervisor to explicitly guide its actions.
- The actions taken by an RL agent have absolutely no influence on the environment.
Which of the following presents a regression task?
Which of the following presents a regression task?
- Segmenting customers into distinct groups based on purchasing patterns.
- Predicting if a patient will develop disease X, given their genome sequence.
- Categorizing customer reviews as positive, negative, or neutral.
- Predicting the float value of the height of a tree, given environmental conditions. (correct)
Which of the following is most clearly a classification task?
Which of the following is most clearly a classification task?
Consider a dataset to fit a linear regression model. What is the predicted value of $y$ at the point $(x_1, x_2) = (0.5, -1.0)$, If (y = \beta_0 + \beta_1x_1 + \beta_2x_2), given the mean-squared error loss, (\beta_0 = 2), (\beta_1 = 1), and (\beta_2 = -1.5)?
Consider a dataset to fit a linear regression model. What is the predicted value of $y$ at the point $(x_1, x_2) = (0.5, -1.0)$, If (y = \beta_0 + \beta_1x_1 + \beta_2x_2), given the mean-squared error loss, (\beta_0 = 2), (\beta_1 = 1), and (\beta_2 = -1.5)?
Consider a k-NN regression model with (k = 3). Which of the following is the predicted (y) value at the point ((x_1, x_2) = (1.0, 0.5)), if the three nearest neighbors have (y) values of 2.65, -2.05, and 1.95?
Consider a k-NN regression model with (k = 3). Which of the following is the predicted (y) value at the point ((x_1, x_2) = (1.0, 0.5)), if the three nearest neighbors have (y) values of 2.65, -2.05, and 1.95?
With a k-NN classifier with (k = 5), you want to predict the class label at point ((x_1, x_2) = (1.0, 1.0)). The five nearest neighbors have class labels: 0, 0, 1, 1, 2. What is the resulting classification?
With a k-NN classifier with (k = 5), you want to predict the class label at point ((x_1, x_2) = (1.0, 1.0)). The five nearest neighbors have class labels: 0, 0, 1, 1, 2. What is the resulting classification?
Regarding linear regression and k-NN regression models, which statement is correct?
Regarding linear regression and k-NN regression models, which statement is correct?
Which of the following options correctly relates model bias and variance to overfitting/underfitting?
Which of the following options correctly relates model bias and variance to overfitting/underfitting?
Model (i) is given by $y = \beta_0 + \beta_1x_1 + \beta_2x_2$ and Model (ii) as $y = \beta_0 + \beta_1x_1 + \beta_2x_2 + \beta_3x_1x_2 + \beta_4x_1^2 + \beta_5x_2^2$. Which of the following options is correct?
Model (i) is given by $y = \beta_0 + \beta_1x_1 + \beta_2x_2$ and Model (ii) as $y = \beta_0 + \beta_1x_1 + \beta_2x_2 + \beta_3x_1x_2 + \beta_4x_1^2 + \beta_5x_2^2$. Which of the following options is correct?
Which of the following correctly describes feature scaling?
Which of the following correctly describes feature scaling?
What is the primary purpose of cross-validation?
What is the primary purpose of cross-validation?
What is the role of the activation function in a neural network?
What is the role of the activation function in a neural network?
What is the learning rate in the context of training neural networks?
What is the learning rate in the context of training neural networks?
What is the purpose of regularization in machine learning?
What is the purpose of regularization in machine learning?
Which of the following is NOT a common distance metric used in k-NN algorithms?
Which of the following is NOT a common distance metric used in k-NN algorithms?
What is the primary goal of anomaly detection?
What is the primary goal of anomaly detection?
Which of the following statements about decision trees is true?
Which of the following statements about decision trees is true?
What is the purpose of a confusion matrix in classification?
What is the purpose of a confusion matrix in classification?
Which of the following describes transfer learning?
Which of the following describes transfer learning?
Flashcards
Unsupervised learning
Unsupervised learning
A learning method where the system learns patterns from unlabeled data.
Identifying close-knit communities
Identifying close-knit communities
Finding clusters or groups within a dataset without predefined labels.
Generate artificial human faces
Generate artificial human faces
Creating new, realistic images from a dataset using generative models.
Reinforcement Learning (RL)
Reinforcement Learning (RL)
Signup and view all the flashcards
RL agents playing against themselves
RL agents playing against themselves
Signup and view all the flashcards
Regression task
Regression task
Signup and view all the flashcards
Predicting CoVID cases and football goals
Predicting CoVID cases and football goals
Signup and view all the flashcards
Classification task
Classification task
Signup and view all the flashcards
Customer repays loan
Customer repays loan
Signup and view all the flashcards
k-Nearest Neighbors (k-NN)
k-Nearest Neighbors (k-NN)
Signup and view all the flashcards
Overfitting
Overfitting
Signup and view all the flashcards
Variance in ML
Variance in ML
Signup and view all the flashcards
Bias in ML
Bias in ML
Signup and view all the flashcards
Overfitting sign
Overfitting sign
Signup and view all the flashcards
Higher variance
Higher variance
Signup and view all the flashcards
Study Notes
- Unsupervised learning problems include:
- Identifying close-knit communities in a social network
- Learning to generate artificial human faces using faces from a facial recognition dataset
- For Reinforcement Learning (RL), the following are true:
- RL agents used for playing turn-based games like chess can be trained by playing the agent against itself (self play)
- RL can be used in an autonomous driving system
- Regression tasks include:
- Predicting the number of new CoVID cases in a given time period
- Predicting the total number of goals a given football team scores in a year
- Classification tasks include:
- Predicting whether or not a customer will repay a loan based on their credit history
- Predicting if a house will be standing 50 years after it is constructed
- Fitting a linear regression model with mean-squared error loss, the predicted value of y at the point (x1,x2) = (0.5, -1.0) is 4.05
- Using a k-nearest neighbor (k-NN) regression model with k = 3 and Euclidean distance, the predicted value of y at (x1,x2) = (1.0, 0.5) is 1.733
- Using a k-NN classifier with k = 5 and Euclidean distance, the class label at the point (x1,x2) = (1.0, 1.0) is 1
- Regarding linear regression and k-NN regression models:
- A k-NN regressor requires the training data points during inference
- A k-NN regressor with a higher value of k is less prone to overfitting
- Regarding bias and variance:
- Bias=E[f^(x)]-f(x); Variance=E[(E[f^(x)]-f^(x))^2]
- Low bias and high variance is a sign of overfitting
- Given two regression models:
- (i) y=β0+β1x1+β2x2
- (ii) y=β0+β1x1+β2x2+β3x1x2+β4x1^2+β5x2^2
- On a given training dataset, the mean-squared error of (i) is always less than or equal to that of (ii)
- (ii) is likely to have a higher variance than (i)
- If (i) overfits the data, then (ii) will definitely overfit
- If (ii) underfits the data, then (i) will definitely underfit
Studying That Suits You
Use AI to generate personalized quizzes and flashcards to suit your learning preferences.
