Symbolic Machine Learning

Semestr: Summer

Range: 2+2


Credits: 6

Programme type:

Study form: Fulltime

Course language: English

Time table at FEE


The course will explain methods through which an intelligent agent can learn, that is, improve its behavior from observed data and domain knowledge. We will be concerned with techniques for the construction of non-trivial symbolic models (graph or logic based) of the observed world. We will explain the basic principles of computational learning theory, which allows to understand why and when successful learning from data is possible. Besides the standard learning scenario involving a prior fixed learning data set, we will discuss other forms of learning, mainly on-line learning, learning through queries (active learning) and reinforcement learning. The class is given in English to all students.


Course syllabus:

1 General framework: observations, actions and rewards
2 Nonsequential and batch learning of concepts, the PAC model
3 Learning selected propositional logic concept classes
4 Learning graphical probabilistic models
5 (lecture 4 cont'd)
6 Learning relational logic concept classes, background knowledge
7 Relational graphical models
8 Learning with queries
9 Active learning
10 Reinforcement learning
11 (lecture 10 cont'd)
12 Relational reinforcement learning
13 Occam priors, Solomonoff induction, Universal AI

Seminar syllabus:

1 Introduction, Python environment, entrance test
2 PEAS agent model, environment properties, input/output types of machine learning models, demos
3 Learning conjunctive and disjunctive concepts
4 Assignment of the first student project
5 Bayesian Networks - semantics
6 Bayesian Networks - inference
7 Assignment of the second student project
8 Inductive Logic Programming - learning from interpretations
9 Inductive Logic Programming - learning from clauses
10 ILP, Q&A
11 Reinforcement learning, demos and introduction
12 Assignment of the fourth project
13 Passive reinforcement learning agents, TD methods
14 Reserve, assessment


Course textbook available at

Stuart Russell and Peter Norvig: Artificial Intelligence: A Modern Approach, Prentice Hall 2010

Luc De Raedt: Logical and Relational Learning, Springer 2008

Marcus Hutter: Universal artificial intelligence, Springer 2005