COMP 5480 - Artificial Intelligence 3

This course provides a comprehensive introduction to the fundamental principles, algorithms, and applications of modern Artificial Intelligence (AI). Designed as a foundational course for students in the MSCS program with an AI concentration, the curriculum covers both classical AI techniques and the core concepts of modern machine learning. The course begins with the formulation of AI problems using the Intelligent Agent paradigm. Students will gain a deep understanding of core AI topics, including uninformed and informed search strategies, adversarial search for game playing, and methods for solving Constraint Satisfaction Problems (CSPs). A significant portion of the course focuses on reasoning under uncertainty, covering the mathematical foundations of probability and their application in Bayesian Networks and Hidden Markov Models (HMMs). The latter part of the course introduces the fundamentals of Machine Learning (ML), including supervised and unsupervised learning paradigms, regression, classification, clustering, and the basic architecture of neural networks and deep learning. Students will develop practical skills by implementing key algorithms in Python and will explore the ethical and societal implications (Fairness, Accountability, and Transparency - FAT) of deploying AI systems. Prerequisite: COMP 5200 Advanced Algorithms Design and Analysis (or equivalent), strong background in linear algebra, probability, and proficiency in Python programming. Prerequisite(s): COMP 5200 Hybrid