3 credits
Spring 2026 Lecture Upper Division
Intelligent Virtual Agents (IVAs) are interactive virtual characters that exhibit human-like qualities, including communicating using natural human modalities such as facial expressions, speech, and gesture. They are capable of real-time perception, cognition, emotion, and action that allow them to participate in dynamic social environments. This course introduces students to the interdisciplinary research on the design, application, and evaluation of IVAs. Students will be taught theory, methodologies, and algorithms, which they will then apply to individual projects. These individual projects will provide students with a common base for authoring IVAs for various application domains, including virtual assistants, pedagogical agents, human-agent task/problem co-solving, human-agent collaboration, and more. Prior background in virtual reality or technical animation is assumed, as the course will provide students the opportunity to further develop these competencies through hands-on assignments within a commercial game engine. The focus of this course is primarily on the computational and functional dimensions of IVAs; however, design and creativity are also important aspects of all projects. This course uses an active learning approach that involves lab assignments, presentations, independent research, and group ideation and collaboration.
Learning Outcomes1Define notions of expressiveness of intelligent virtual agents.
2Create domain-specific IVAs using artificial intelligence techniques.
3Identify application domains and simulate IVAs' behaviors.
4Develop communication strategies that enable human-IVA interaction based on artificial intelligence techniques.
5Apply the theory and programming knowledge covered in class to concrete exercises using a commercial game engine.
6Compare and contrast the constraints and affordances for expressiveness in physical vs. virtual intelligent agents and human-like vs. non-humanoid agents.
Restrictions
made by Purdue students.