CSC3203 : Artificial Intelligence for Games

  • Module Leader(s): Dr John Shearer
  • Owning School:
Semesters
Semester 1 Credit Value: 10
ECTS Credits: 5.0

Aims

• To introduce the concepts and practices of artificial intelligence.
• To develop an awareness of the use and potential use on artificial intelligence in computer game applications.
• To develop an understanding and gain practical experience of the implementation of artificial intelligence algorithms with the context of computer games.

Games were among the first application areas of Artificial Intelligence (AI). For decades researchers tried to employ AI techniques to challenge human players in a wide variety of games. The success of IBM's Deep Blue in beating world champion Gary Kasparov at chess has shown that computers can eventually become better than the human players at games that have been considered a measure of intelligence. The module will present highlights of the past, latest developments in the present, and some outlook to the future of AI in game playing. Among the methods presented in the course are classic search algorithms, planning methods, agent-based AI, and machine learning. Practical classes will provide an insight into the design, implementation and evaluation of artificial intelligence techniques.

Outline Of Syllabus

• Introduction to artificial intelligence
• Finite state machines and games
• Problem solving as search
• Heuristic search
• Pathfinding
• Game playing as search
• Machine learning for formal games
• Knowledge representation
• Planning
• Planning and acting in the real world
• Learning from observations
• Statistical learning methods - evolutionary optimisation and multi-agent systems
• Design and development of artificial intelligence algorithms in Java
• Game AI and the agent paradigm
• Wider considerations when designing an AI for a game

Teaching Methods

Teaching Activities
Category Activity Number Length Student Hours Comment
Scheduled Learning And Teaching ActivitiesLecture161:0016:00Lectures
Guided Independent StudyAssessment preparation and completion200:3010:00Revision for end of Semester exam and exam duration
Guided Independent StudyAssessment preparation and completion161:0016:00Lecture follow-up
Scheduled Learning And Teaching ActivitiesPractical111:0011:00Practicals
Guided Independent StudyProject work111:0011:00Coursework
Guided Independent StudyIndependent study361:0036:00Background reading
Total100:00
Teaching Rationale And Relationship

Lectures will be used to introduce the learning material and for demonstrating the key concepts by example. Students are expected to follow-up lectures within a few days by re-reading and annotating lecture notes to aid deep learning.

This is a very practical subject, and it is important that the learning materials are supported by hands-on opportunities provided by practical classes. Students are expected to spend time on coursework outside timetabled practical classes.

Students aiming for 1st class marks are expected to widen their knowledge beyond the content of lecture notes through background reading.

Students should set aside sufficient time to revise for the end of semester exam.

Assessment Methods

The format of resits will be determined by the Board of Examiners

Exams
Description Length Semester When Set Percentage Comment
Written Examination901A50N/A
Other Assessment
Description Semester When Set Percentage Comment
Practical/lab report1M15uninformed search coursework (5 hours)
Practical/lab report1M35Informed search coursework (20 hours)
Assessment Rationale And Relationship

Since artificial intelligence is in part a practical subject, the design and implementation of software incorporating artificial intelligence techniques need to be assessed by coursework. However, students must also demonstrate an understanding of the key theoretical issues and be able to apply this understanding which will be achieved in a formal examination.

N.B. This module has both an “Exam Assessment” and an “Other Assessment” (e.g. coursework). If either Assessment has a failing mark below 35%, the maximum mark returned for the module will normally be 35%.

Reading Lists

Timetable

Disclaimer: The University will use all reasonable endeavours to deliver modules in accordance with the descriptions set out in this catalogue. Every effort has been made to ensure the accuracy of the information, however, the University reserves the right to introduce changes to the information given including the addition, withdrawal or restructuring of modules if it considers such action to be necessary.