This page is under construction
Syllabus
Fall Semester, 2026
Classroom: TBA
Time: TBA
Instructor: Dr. Michael Weeks
Computer Science Department
Office: 25 Park Place, room 754
office hours may be conducted via Webex as well as in-person.
Office Hours: TBA
web-page: http://hallertau.cs.gsu.edu/~mweeks
Teaching Assistant: TBA
TA's office hours: TBA
TA's office: office hours will be held via Webex
TA's e-mail address: TBA [at] gsu [dot] edu
Click here for the Syllabus policies
FINAL EXAM
The Final Exam will be presentations, held on
TBA, from TBA to TBA
(the official final exam time).
DESCRIPTION
Advanced AI algorithms and tools used in gaming; topics include
genetic algorithms and neuroevolution, Monte-Carlo tree search,
finite state machines,
procedural content generation,
path finding, agents, and
reinforcement learning.
TEXTS
- Selected research papers.
PREREQUISITES
CSc 3320.
Programming maturity is assumed.
In addition, students are expected to know
discrete structures applicable to computer science, number bases,
logic, sets, Boolean algebra, graph theory.
CONTENT
This course discusses research papers related to AI in games,
including
genetic algorithms and neuroevolution, Monte-Carlo tree search,
finite state machines,
procedural content generation,
path finding, agents, and
reinforcement learning.
Students will read research papers, present them, answer questions about them, and review each other's presentations.
GRADING
- Participation and Attendance (and paying attention) will constitute 5% of the course grade.
- A mid-term exam will constitute 20% of the course grade.
- Any pop-quizzes will factor into the exam grade at 1/10 the weight of the exam.
- Approximately 4 Assignments will constitute 40% of the course grade. (There will be at least 3 assignments, maybe as many as 6). Assignments will include a paper summary, and a literature review. These may be in several parts or forms, such as requiring a written paper, an in-class presentation, a question-answer session, and potentially a video.
- The project will constitute 35% of the course grade. This includes several reports and/or in-class presentations (e.g. a project proposal, a mid-semester update, and a final presentation).
LEARNING OUTCOMES
By the end of this course, students will know:
Fundamental Concepts
- Analyze and explain the role of AI in video games, including its history and impact on game design and player experience.
- Understand and apply foundational AI concepts such as state-space representation, heuristic search, and decision-making processes.
- Implement and evaluate various pathfinding algorithms like A*, Dijkstra's, and Breadth-First Search to enable non-player characters (NPCs) to navigate game environments efficiently.
- Develop and critique steering behaviors (e.g., seek, flee, flocking) for realistic and dynamic character movement.
- Design and create finite-state machines (FSMs) and behavior trees to model complex NPC behaviors, from simple patrols to intricate combat strategies.
- Utilize utility-based AI to enable NPCs to make intelligent, context-aware decisions in dynamic situations.
- Apply and evaluate machine learning techniques, such as Q-learning and neural networks, to enable NPCs to learn from player interactions and adapt their behaviors.
- Train a simple game AI to play a classic game (e.g., Tic-Tac-Toe, Pac-Man) using reinforcement learning.
- Develop and implement systems to model player behavior and preferences, allowing for dynamic difficulty adjustment and personalized content.
- Create procedural content, such as generating levels, quests, or textures, using AI algorithms.
- Integrate multiple AI algorithms to create a complete and compelling AI for a game prototype.
- Debug and optimize AI code to ensure efficient and performant behavior within a game engine.
- Communicate and present their AI design and implementation choices effectively.