## Q&A Sessions

The weekly Q&A sessions serve two goals. Firstly, Big Data is a rather theoretical course. The math is not deep, but some of the concepts may be new to you and the many theorems and proofs may seem daunting. So, whenever you have a question during the lectures you should ask. But then it can still happen that you revise the material afterwards and a new question comes up. That is what these Q&A sessions are for. To make it even more convenient, each Q&A session in Teams has a document in which you can enter your questions whenever they come up.

Secondly, to get some practice with the theoretical concepts discussed during the lectures, exercises and their solutions are provided. Clearly, you should try to solve the exercises yourself before you look at the provided answers! All questions regarding these exercises can be asked in the same way as your questions on the lectures. Note, all exercises and solutions have been provided by dr A.J. Feelders (many thanks!).

Every week, we will discuss all the questions you have entered as well as any other questions that you pose "live".

- The first Q&A Session is on Friday, February 11. As a preparation for this session, please watch Lecture 1 (Probability Models and Axioms), video's 1 through 13 (De Morgan's Laws) of the MIT course Introduction to Probability. Total running time is about 1.5 hours.

### Schedule

Week | Subject/Exercises | Preparation |
---|---|---|

6 | Introduction to probability Exercises (Solutions) |
Introduction to Probability:
Lecture 1 (Probability Models and Axioms), video's 1 through 13 (De Morgan's Laws) |

7 | Introduction to probability (continued) Exercises (Solutions) |
Introduction to Probability:
Lecture 2 (Conditional Probability), Lecture 3 (Independence), Lecture 4 (Counting), Lecture 5 (Random Variables) |

8 | Sampling for frequent item set mining Exercises (Solutions) | Lecture slides of lecture 3 and 4 |

9 | PAC learning with finite hypothesis classes Exercises (Solutions) | Lecture slides of lectures 5 and 6 |

10 | Infinite hypothesis classes and the VC-dimension Exercises (Solutions) | Lecture slides of lectures 7 and 8 |

11 | PAC based sampling for frequent item set mining Exercises (Solutions) | Lecture slides of lectures 9 and 10 |

12 |