Overview
This course explores the latest breakthrough in Large Language Models and the emergence of Generative Agents, showcasing how they can simulate human-like behavior in an interactive sandbox. The learning outcomes include understanding the three key components of Generative Agents and how they work together to plan behavior. The course teaches about the technology that extends Language Models to store a complete record of an agent's experiences using natural language and synthesize those memories over time into higher-level reflections. The teaching method involves a video presentation with sections covering the introduction, Generative Agents Paper, agent goals, agent loop, Smallville Sandbox World, and conclusion. The intended audience for this course includes individuals interested in artificial intelligence, machine learning, and the simulation of human behavior.
Syllabus
- Intro
- Twitter Source
- Generative Agents Paper
- Agent Goals
- Agent Loop
- Smallville Sandbox World
- Conclusion
Taught by
Venelin Valkov