PSU Cmpsc 442 - Artificial Intelligence, Spring 2023

Tues & Thurs 4:35-5:50pm, 104 Thomas

Instructor: Prof. Chris Dancy

How can we build complex information processing systems? What do we need to represent and how do we represent it? How can we make such systems learn, and perceive in an environment? How might we use the way people think to design these systems? What about the environment, how should it affect an agent? How do societal and sociocultural structures provide foundations for the AI systems we create, deploy, and integrate?

In this course, I will give you some tools to provide some answers to these questions. You will have an opportunity to explore past answers to these questions and learn from them. Furthermore you will have an opportunity to use the thinking, techniques, and tools you learn in this course and in the future!

This course will not provide full coverage of Artificial Intelligence for this would almost certainly be impossible given the time! These fields are big with many theories, subfields, and applications. This course will focus on a balance between philosophical, historical, social, cognitive, and methodological aspects of AI.

We will learn about the following topics:

Historical perspectives on AI
Neural Networks
Trees, Searching them, and Using them
Reinforcement Learning
AI & Society
RBES & Cognitive systems

As we cover these topics we’ll also critically consider the ethical and social sides of these techniques applied on various data as well as the same ethical and social sides of construction of those systems and data.

Course Outcomes

• Students will be able to explain the fundamentals of some AI tools and techniques, and implement an technique/tool used to build AI/Complex Information Processing Systems.
• Students will understand some of the historical perspectives of AI be able to think critically about these perspectives
• Students will understand what agents and models are, particularly as they apply to building AI/complex information processing systems.
• Students will have sufficient opportunity to contribute to a scientific style report that accompanies a software implementation of a technique/algorithm/theory, including documentation for such artifacts.