Crowdly
Add to Chrome
COMP9417-COMP9417-Machine Learning & Data Mining - 2026 T2
Looking for COMP9417-COMP9417-Machine Learning & Data Mining - 2026 T2 test answers and solutions? Browse our comprehensive collection of verified answers for COMP9417-COMP9417-Machine Learning & Data Mining - 2026 T2 at moodle.telt.unsw.edu.au.
Get instant access to accurate answers and detailed explanations for your course questions. Our community-driven platform helps students succeed!
Consider the RBF kernel from lectures . Suppose we have three points , and you are told that geometrically, is very close to , and is geometrically far away from . Which of the following statements are true?
A. is close to , and is close to
B. is close to , and is close to
C. is much larger than , and is much smaller than
D. is much smaller than , and is much smlargerller than
Your dataset consists of documents, each of which may be represented as a 3 dimensional feature vector. You decide to fit a logistic regression to the data, and derive the following estimates for your weight vector: . You then receive a new document . Compute .
Consider the following dataset with features X, Y and Z
| X | Y | Z |
| 0 | 0 | 0 |
| 0 | 1 | 1 |
| 1 | 0 | 0 |
| 1 | 1 | 1 |
| 2 | 0 | 1 |
| 2 | 1 | 1 |
Compute the entropy of
Which of the following statements is/are CORRECT about ensemble learning? (you can choose more than one option)
You have two classes of 1 and -1 and you run AdaBoost with a logistic regression base classifier for three iterations. After doing so, you have three logistic regression models which achieve weighted errors respectively. For a new test point, the predictions of your three models are respectively. What is the prediction of your boosted model when you use a threshold of 0.6 for the base models?
Consider the following dataset: with corresponding labels . Which of the following transformations would make the data linearly separable?
A.
B.
C.
D.
A multilayer perceptron consists of an input of size 5, fully connected to a hidden layer of size 20, fully connected to an output of size 3.
Given a bias is used at each hidden node and each output node, what is the total number of parameters that need to be learned for this network?
Which of the following models can NOT achieve training error of zero on any linearly separable dataset? (you can choose more than one option)
100 individuals that are known to have COVID19 are tested for the virus using a new take-home testing kit, and of those 100, only 72 test positive.
Let and and let kernel k be defined as follows:
which transformation does this kernel correspond to?