Department of Information and Computing Sciences

Departement Informatica Onderwijs
Bachelor Informatica Informatiekunde Kunstmatige intelligentie Master Computing Science Game&Media Technology Artifical Intelligence Business Informatics

Onderwijs Informatica en Informatiekunde

Vak-informatie Informatica en Informatiekunde

Methods in AI Research

Course code:INFOMAIR
Credits:7.5 ECTS
History:This course was formerly known as Methods in AI research (INFOMMAIR). You can only do one of these courses.
Period:period 1 (week 36 through 45, i.e., 3-9-2018 through 9-11-2018; retake week 1)
Participants:up till now 108 subscriptions
Schedule:Official schedule representation can be found in Osiris
lecture   Tue 13.15-15.0036 UNNIK-GROEN Floris Bex
Kees van Deemter
Thu 11.00-12.4536 RUPPERT-BLAUW
37-44 RUPPERT-042
tutorial group 1 Tue 15.15-17.0037-44 RUPPERT-C Marijn Schraagen
studentassistent BS
group 2 Tue 15.15-17.0037-44 RUPPERT-033 Onuralp Ulusoy
group 3 Tue 15.15-17.0037-44 RUPPERT-B
week: 1Thu 3-1-20199.00-12.00 uurroom: EDUC-BETAretake exam
Contents:Artificial Intelligence is a fast-paced and challenging field that is making visible inroads into our everyday life. AI in Utrecht offers a unique integrated and modern approach, studying AI from the viewpoints of Informatics, Logic, Cognition, Psychology, Philosophy, and Linguistics. Because of its interdisciplinary character, the variety of techniques and methods used in AI is considerable, ranging from theoretical to empirical, and from formal mathematical to more informal. In this course, we will introduce the various perspectives on AI and the methods associated with them, with a particular emphasis on perspectives in Utrecht. We will discuss a brief history and overview of AI as well as models of cognition, communication, knowledge, learning, and language, and the ways in which these models can be evaluated and ultimately used to generate intelligent behavior. The linking pin of the course will be a central lab project in which you will develop, describe, test, and evaluate a dialog system (sometimes also referred to as “chatbot”). In this way, the theory from the lectures forms the basis of a real AI application. In the course, there are two lecture sessions of 1.5 hours per week. The following topics will be discussed during the lectures. * Dialogues & Dialogue Systems * Machine Learning & Agent Learning * Intelligent Agents * Computational Argumentation * Cognitive Modeling * Experimentation: user studies, system evaluation & brain and behavior * Logic Modelling & Logic Tools * Logic & language * Models of Language * Linguistic Interaction * Natural Language Generation There is also a lab session of 1.5 hours per week. In the labs, you will work on one central project in groups of 4, namely a working chatbot/dialog system. The system should have the following components: * Domain modelling - Determine which goals the dialog system will support. * User input processing - Given user input in natural language, apply machine learning and keyword-based pattern matching rules for classifying this input. * Reasoning - Given the processed input, determine a suitable system response by reasoning with structured knowledge. * Generating responses - Generate responses in natural language. Furthermore, in addition to the system a scientific report will have to be delivered that includes: * Description of the system. * Technical evaluation - how does the final system compare to your initial objectives from a technical perspective? What are limitations and how would you tackle these in future work? * User evaluation - how does the final system compare to your initial objectives from a user perspective? What do users think of the system?
Literature:See Blackboard page for the course.
Course form:Lectures and lab sessions.
Exam form:Written exam (50%) and system development (50%)
Minimum effort to qualify for 2nd chance exam:Om aan de aanvullende toets te mogen meedoen moet de oorspronkelijke uitslag minstens 4 zijn.