CS 493/593 Special Topics: Conversational and Agentic AI
(Fall 2025, Spring 2027)
This course explores the design of intelligent agents, focusing on computer systems that reason and interact through language. The course builds from foundations of goal-oriented dialogue systems to current research in Large Language Model Agents that execute perception-action loops in digital and physical environments. Topics include:
- language modeling, from underlying representations like word vectors to state-of-the-art models like generative pre-trained transformers (GPTs);
- planning in conversations, tracing methods from early computational pragmatics to modern techniques like reinforcement learning (RLHF, RLVR, RLCD);
- current agent architectures, covering how LLM agents can execute complex plans, maintain memory, use tools, and integrate perception and action for tasks in physical environments; and,
- applications such as web agents and physical AI (robots, autonomous vehicles, etc.)
CPE 520: Applications of Neural Networks
(Fall 2026, Fall 2027)
This course provides an introduction to modern neural networks, with an emphasis on understanding how these models work and how they can be applied to practical problems. The course begins with foundational machine learning concepts, including loss functions, optimization, evaluation, and generalization. It then builds on these foundations to cover dense neural networks, convolutional and recurrent networks, attention and transformers, representation learning, and selected topics in neural network learning theory. The course emphasizes practical applications while introducing the theoretical concepts necessary to understand model behavior and make informed decisions in application settings. By the end of the course, students should be able to implement neural networks, assess the suitability of different architectures for a given problem, evaluate model performance appropriately, and interpret current research literature to adapt relevant methods to new applications.
CS 220: Discrete Mathematics
(Spring 2026)
Discrete mathematics provides the formal language behind algorithms, data structures, and system design. It is fundamental to traditional computing systems and modern artificial intelligence systems alike. This course develops the reasoning and modeling tools needed to analyze and justify how computational systems behave. Topics include propositional and predicate logic, mathematical proofs, sets and tuples, binary relations and graph representations, induction and recursion, and combinatorial counting. Emphasis is placed on using these concepts to model discrete structures, establish mathematical properties, and reason rigorously about computational problems.
CSEE 480/481: Capstone Design and Implementation
(Fall 2026, Spring 2027)
This capstone explores the design and implementation of intelligent agents that reason and act in dynamic environments. Students develop an AI agent capable of interpreting natural-language instructions and controlling a character in ET: Legacy, a real-time multiplayer game. The project draws on concepts from large language models, planning, perception, memory, and feedback-driven decision making. Emphasis is placed on integrating these components into a working system that can translate high-level strategies into executable behavior, adapt to changing conditions, and operate under real-time computational constraints. Students evaluate their agents through empirical testing and competitive gameplay.