Blog

Lesson 1 k-Nearest Neighbors   Let assume that a dataset consists of 4 data points. The followings are the coordinates and the label of each data point:   Point A: (2, 4) / class 0 Point B: (0, 0) / class 1 Point C: (-1, 3) / class 0 Point D: (3, 2) / class 1   If you train a 1-NN model on this dataset, what would be the prediction for Point B?

Leeson 1 Overview and Normal Distributions   What do generative models generate?

Lesson 3 Kernel Density Estimation   Which of the following is true about the kernel density estimation?

Lesson 4 Derivation of Discriminative Loss Function   You are classifying the binary dataset of 300 data points. Which of the following cannot be the value of a counting loss function?

Lesson 1 Perceptron Model   The max cost function g(w) takes a 1×3 parameter vector as an argument. Which parameter vector results in the best separation?

Lesson 1 One-hot Encoding   Let’s predict the grade of this course using the following features:   – Class section (categorical): 4745, 6745-Q, or 6745-AQ – Midterm score (numerical): 0-100 – Project group (categorical): Group 1, 2, 3, 4, 5, 6, 7, and 8 – Final exam score (numerical): 0-100   After applying one-hot encoding to all categorical features, the total number of features will be

Lesson 1 Overview of Python   Which of the following statements is not correct about the Python language?

Lesson 4 Outliers   Which of the following is not true about the z-score approach?

Lesson 1 Simple Linear Regression   Which of the following is not true?

Lesson 3 Kernel Regression   Assume that a training set has 300 data points and the test set has 100 data points. What would be the dimensions of an RBF kernel matrix used to predict the training set?