DEPARTMENT OF ENGINEERING MATHEMATICS

EMATM1100

Knowledge Representation & Uncertainty

 

(10 credits)

Timetable Assessment  Syllabus Materials Textbooks Past exams

Organiser:

Dr E Di Tomaso

Lecturer:

Dr E Di Tomaso

Description:

 This unit presents the fundamentals of logical systems and probabilistic reasoning systems for knowledge representation and uncertainty.
Pre-requisites: EMAT31600 Computational Intelligence
Aims: To give a state-of-the-art presentation of methods used in Artificial Intelligence for the representation of knowledge and the management of uncertainty.

Learning outcomes:

 On successful completion of this unit, students will:
  • understand two key areas underlying the design and implementation of knowledge-based systems, namely knowledge representation and uncertainty
  • develop a critical awareness of the relative advantages and disadvantages of different schemes.

Organisation & timetable:

Timetabling to be confirmed.

Assessment:

 2 hours written examination.

Syllabus:

 Knowledge and Reasoning
Logical Agents, Inference in First-Order Logic;
Non-standard Logics:   Non-monotonic Logics, Modal Logic;
Specialised Reasoning Systems: Semantic Networks, Conceptual Graphs.

Uncertain Knowledge and Reasoning
Uncertainty;
Probabilistic Reasoning: Bayesian Networks, Exact and Approximate Inference;
Probabilistic Reasoning over Time: Inference in Temporal Model, Probabilistic Temporal Models;
Decision Problems under Uncertainty: Utility Theory, Decision Networks, Dynamic Decision Networks;
Uncertain Reasoning from Observations: Statistical Learning Methods.

Materials: Handouts

Books:

 "Artificial Intelligence: a Modern Approach" by Stuart Russell and Peter Norvig (Second Edition),  Prentice Hall 2003
Past exams: /private/teaching/local/exam-p-and-s/exams-index.html