General information

• Study points: 2.00
• Language: English
• Category: Assignments
• Duration: 12.0 hours 
• Periodicity: Taught in the first semester
• POC: POC Artificial Intelligence

Taught by

Blockeel Hendrik

Aims

Content

The students will use the knowledge gained from the lectures and exercise session to build a practical system that exhibits learning behavior.

Course Material

A project description, possibly with pointers to relevant literature.

Course activities

The project runs from week 4 until week 13 of the second semester.

• In week 4, students receive a description of a program that is to be implemented, with instructions or suggestions as deemed necessary.  Students will typically work on the project in groups of 2-3 people.
• Around week 8, a brief report is expected. This report is intended to allow the teaching assistants to provide feedback when necessary.
• In week 11, a final report and software is to be handed in.
• Around week 13, a brief discussion of the project results is held, and final feedback is given to individual groups.

General information

• Study points: 1.00
• Language: English
• Category: Exercises
• Duration: 15.0 hours 
• Periodicity: Taught in the first semester
• POC: POC Artificial Intelligence

Taught by

Blockeel Hendrik

Content

Exercises are made on the subjects discussed during the lectures. These are mostly pen-and-paper exercises where students gain insight in the workings of learning algorithms by manually mimicking the computations of certain learning algorithms, graphically describing the result of a learning algorithm (by drawing decision
surfaces), etc.  There are also exercises on evaluation of machine learning models and algorithms.

Course Material

• A list of exercises.
• Solutions are made available on Toledo.

Course activities

Students try to independently solve the exercises during some time.  A teaching assistant provides help where necessary, and discusses the solution afterwards.

This course is also included in

H02C1A Machine Learning and Inductive Inference  

General information

• Study points: 3.00
• Language: English
• Category: Lectures
• Duration: 19.5 hours 
• Periodicity: Taught in the first semester
• POC: POC Artificial Intelligence

Taught by

Blockeel Hendrik

Content

1. introduction to machine learning, connections with other subjects
2. general principles of learning:
- concept learning, version spaces
- evaluation of learning algorithms
- theory of learnability
- representation of inputs and outputs of learning algorithms
3. specific learning approaches:
- decision trees
- rules, association rules
- instance based learning
- clustering
- neural networks
- support vector machines
- Bayesian learning
- genetic algorithms
- ensemble methods (bagging, boosting, ...)
- reinforcement learning
- inductive logic programming
 

Course Material

Course Text
Lecture slides

Course activities

Ten lectures of 2 hours each.

This course is also included in

H02C1A Machine Learning and Inductive Inference  

Evaluation description

Examination type:
    oral with written preparation
When?: final examination during examination period
Evaluation type:
    Open book
    Report

Explanation

The exam consists of questions about the theory as well as some exercises.  The course text and lecture slides can be consulted during the exam.  Students have about 2 hours time for preparing their answers, followed by an ten minute oral interrogation.

The project influences the final score for the course.  With
P: score on the project (0-5)
E: score on the exam (0-20),
F: final score
the final score is computed as follows:
F = 3/4 E + P if P>0
F = 0  otherwise
That is: the exam score is rescaled to a maximum of 15 and the project score is added, with the exception that a student cannot pass the exam with a 0 score on the project.
For the retake in September, the exam format is the same.  If the project was considered insufficient,
the student is given the opportunity to improve it according to concrete suggestions.