COMPUTER SUBJECT:                                                           BASIC ML CONCEPTS

 

TYPE:                                                         GROUP WORK EXERCISE/DISCUSSION

 

IDENTIFICATION:                          HOMEWORK1/MICL

 

COPYRIGHT:                                     Michael Claudius

 

LEVEL:                                                            EASY

 

DURATION:                                         1-3 hours

 

SIZE:                                                     100 lines!!

 

OBJECTIVE: Introduction to ML basic

 

REQUIREMENTS:                             ML Ch. 1

 

COMMANDS:                                     

 


IDENTIFICATION: HOMEWORK1/MC

 

 

ML Chapter 1 Assignments

 

The following assignments must be solved in groups (4-5 persons) and followed up by a short presentation in the class later.

 

1.                   How would you define Machine Learning?

2.                   Can you name four types of problems where it shines?

3.                   What is a labeled training set?

4.                   What are the two most common supervised tasks?

5.                   Can you name four common unsupervised tasks?

6.                   What type of Machine Learning algorithm would you use to allow a robot to walk in various unknown terrains?

7.                   What type of algorithm would you use to segment your customers into multiple groups?

8.                   Would you frame the problem of spam detection as a supervised learning problem or an unsupervised learning problem?

 

9.                   What is an online learning system?

10.                 What is out-of-core learning?

11.                 What type of learning algorithm relies on a similarity measure to make predic‐tions?

12.                 What is the difference between a model parameter and a learning algorithm’s hyperparameter?

13.                 What do model-based learning algorithms search for? What is the most common strategy they use to succeed? How do they make predictions?

14.                 Can you name four of the main challenges in Machine Learning?

15.                 If your model performs great on the training data but generalizes poorly to new instances, what is happening? Can you name three possible solutions?

16.                 What is a test set, and why would you want to use it?

17.                 What is the purpose of a validation set?

18.                 What is the train-dev set, when do you need it, and how do you use it?

19.                 What can go wrong if you tune hyperparameters using the test set?