Syllabus and course materials for the AI, Games, and Design (ECS 289G 002 for fall quarter 2020).
At the intersection of computation, games, and design lies interdisciplinary problems that present some of the most difficult challenges facing artificial intelligence. Currently, most of AI's application to games is in the form of playing games such as AlphaStar playing StarCraft II and Counterfactual Regret Maximization solving heads-up limit hold'em poker. When the AI system is applied to design, the supporting structures such as error minimization, reward functions, and easily-quantized vector-based representation are replaced by the contextual and subjective nature of design problems.
This course is about the application of AI systems into game design to enhance existing games and to explore previously unreachable design spaces. Topics will include interactive storytelling, social characters, procedural content generation, and modes of interaction. The AI techniques covered are broad and may include machine learning, logic-based inference, planning, and search. Foundational and open-ended topics of AI will be discussed including knowledge representation, contextual reasoning, and the integration of approaches in AI (e.g. systems that are both statistical learnings and symbolic reasoners). Assessment will consist of a term project, smaller technical projects, and responses or reviews of research papers, videos, games, and other media.
Meeting times: 9:00-10:20 AM PST on Tuesdays and Thursdays
Office Hours: Tuesdays 10:30-11:30 AM PST, Wednesdays 3:00-4:00 PM PST, private meetings scheduled as needed
Media for Class Meetings and Office Hours: Zoom
Lectures will be recorded.
Holidays: Nov. 11th, Veterans Day (no office hours); Nov. 26th, Thanksgiving (no class meeting)
Scheduled Finals Times: Monday, December 14th at 10:30 AM-12:30 PM PST
Information: UC Davis 2020 Academic Calendar | Fall 2020 Finals Schedule
Dr. Joshua A. McCoy, Assistant Professor
Computer Science and Cinema and Digital Media Departments
Email: jamccoy at ucdavis dot edu
Office: 3033 Kemper Hall
http://joshmccoy.com
https://faculty.engineering.ucdavis.edu/mccoy/
Zee Chen
One of Josh's Ph.D. students working in computational media with an interest in expressive AI techniques for interactive characters using methods from academic AI, machine learning, and industry games AI techniques.
Feel free drop me a message on Discord or contact me by email: zeychen at ucdavis dot edu
- [30] Reading Responses
- [30] Programming Projects
- [40] Final Project
The goal of the reading responses is to assess how students are absorbing the course materials and how able they are to synthesize the concepts into statements about the topic area. As this course is virtual and does not have a discussion section, receiving critical feedback on responses is a mechanism that facilitates student-to-student idea exchange. This instructor hopes that having your thoughts on a course read and responded to by peers will provide compelling and valuable feedback.
There will be approximately 8 pairs of written responses. The first of the pair wil be a 500 word response based on the course material and a writing prompt. The second of the pair will be a 250 word reaction and critique of another student's first response. The instructor and TA will map pairs of students for the second response of the pair.
Reading responses are due 1 week after it's assigned. Critiques are due 1 week after the responses are due.
Check Reading Response Assignment for weekly assignments. Each sheet corresponds to the week the reading response is assigned.
If you would like to use a personalized username for the reading responses, please fill out the RR User Name Form.
There will be 3 programming projects in this course that when completed will provide a baseline of game AI techniques for you to construct a final project from. These projects are designed to introduce you to valuable frameworks and systems used in AI-based game design. The tentative list of projects are:
- Classic non-player character behavior controllers.
- Procedural story generation via natural language generation algorithms.
- Social and story modeling with Comme il Faut/Ensemble.
You will be responsible for the creation of prototype AI-based game. You can work in groups of 1-3 students. The project will involve a written proposal, project update presentaitons, and final presentation, and a repository featuring a playable demo or video of your prototype.
The UC Davis Code of Academic Conduct (http://sja.ucdavis.edu/files/cac.pdf) will be strictly enforced in this class. In particular, plagiarism, academic dishonesty, and cheating will be dealt with severely. Any breach of the Code of Academic conduct can result in failing the assignment, failing the course, and displinary action via the Office of Student Support and Judicial Affairs (http://sja.ucdavis.edu/).
The use of laptops and technology in general are encouraged in this course as long as they are not disruptive to the rest of the class. If you choose to use a device with a screen, please sit in the back of the room to avoid distracting your fellow students. You are required to ask for permission before video or audio recording in the classroom. In general, students will be treated as adults capable of managing their technological lives while being respectful of others in the classroom.
Students are not permitted to make visual or audio recordings, including live streaming, of classroom lectures or any class related content, using any type of recording devices (e.g., smart phone, computer, digital recorder, etc.) unless prior permission from the instructor is obtained, and there are no objections from any of the students in the class. If permission is granted, personal use and sharing of recordings and any electronic copies of course materials (e.g., PowerPoints, formulas, lecture notes and any classroom discussions online or otherwise) is limited to the personal use of students registered in the course and for educational purposes only, even after the end of the course.
To supplement the classroom experience, lectures may be audio or video recorded by faculty and made available to students registered for this class. Faculty may record classroom lectures or discussions for pedagogical use, future student reference, or to meet the accommodation needs of students with a documented disability. These recordings are limited to personal use and may not be distributed (fileshare), sold, or posted on social media outlets without the written permission of faculty.
Unauthorized downloading, file sharing, distribution of any part of a recorded lecture or course materials. or using information for purpose other than the student's own learning is prohibited unless prior authorization is given by the instructor.