Agent-based Modeling and Agentic Technology
94-815
Units: 6
Description
No pre-requisites
As AI rapidly evolves, the integration of Large Language Models (LLMs) is transforming agent-based modeling, creating agents with logical reasoning, behavior modeling, contextual adaptation, and real-time information retrieval. Recognized by analysts and venture capital firms as a key technology trend, Agentic AI enables autonomous systems to manage workflows and support strategic decision-making across industries, reshaping single-agent and multi-agent systems.
This course on Agent-Based Modeling and Agentic Technologies offers an in-depth exploration of systems thinking, simulation techniques, and LLM-powered agents. Students will develop foundational knowledge in agent-based modeling and progressively engage with LLM-driven agents, capable of complex interactions and adaptive responses. Applications span fields such as scientific research, policy simulations, and industrial automation, enabling students to deploy autonomous, reasoning-enabled agents in diverse scenarios.
Blending theoretical insights with hands-on applications, this course prepares students to apply agentic AI in domains like business process management, healthcare, and environmental science, where adaptability and strategic decision-making are essential. Through industry scenarios, case studies, and guest lectures, students will discover how agentic technologies enhance workflows and decision-making, empowering them to lead AI-driven innovations in their fields.
Learning Outcomes
Upon completion of this course, students will be able to:
- Understand and apply systems thinking principles to model complex systems using agent-based and other simulation methodologies (e.g., Monte Carlo, discrete event simulations).
- Analyze and construct foundational agent-based models (ABMs) that incorporate agent interactions and environmental dynamics.
- Explain the architecture, capabilities, and applications of Large Language Models (LLMs) in the context of agentic technologies and their role in enhancing agent behaviors.
- Design and deploy LLM-based agents using core frameworks (e.g., LangChain, AutoGen), implementing advanced data retrieval, prompt engineering, and agent interaction strategies.
- Differentiate between various types of LLM-based agents, such as reasoning, decision-making, and experiential learning agents, and select appropriate types for specific applications.
- Evaluate the real-world applications of agent-based and LLM-based agents across diverse sectors, including strategic decision-making, autonomous robotics, and infrastructure management.
- Develop fine-tuning and validation strategies for multi-agent systems, ensuring robust, role-specific performance and inter-agent coordination.
- Assess deployment, optimization, and governance challenges associated with scaling LLM-based agents, focusing on performance, ethics, and industry best practices.
Prerequisites Description
No pre-requisite courses. Introduction to analytics, data science, and generative AI would be useful.