CS 2510 Artificial Intelligence - Foundations and Currents

Author

Affiliation

Clayton Cafiero

University of Vermont

In 2024 fall and 2025 spring this was taught as a topics course as CS 2990. Added to permanent course catalogue starting fall 2025 as CS 2510. Soon to be renamed: Artificial Intelligence: Foundations and Currents

Prerequisites: CS2240 Data Structures and Algorithms

Topics:

• History of artificial intelligence
• Logic programming with Prolog
• Knowledge representation and ontologies
• Heuristic search, A^{\star} and I\mkern-2muD\mkern-2.3muA^{\star}
• Adversarial games, minimax, expectiminimax, \alpha–\beta pruning, horizon problem
• Markov models and Markov decision processes
• Decision trees, random forest, and ensemble learning
• Introduction to neural networks (multi-layer perceptron)
• Current state of AI; implications for humanity and society
• Other topics TBD

Introduction and brief History

• Introduction and brief history

Logic review

• Boolean logic
• Propositional logic
• First-order predicate logic
• Proof systems
• Horn clauses

Logic programming

• Prolog primer
• Resolution and backtracking
• Prolog and Horn clauses
• Towers of Hanoi
• Knowledge representation and ontologies (forthcoming)
• Expert systems (forthcoming)

Heuristic search: A^{\star} and I\mkern-2muD\mkern-2.3muA^{\star}

• For review of Dijkstra’s algorithm, see materials for CS 2240.
• Heuristic search
• A^{\star}
• Dijkstra’s algorithm and early termination
• Admissibility and consistency
• A few heuristics
• Choosing a heuristic
• I\mkern-2muD\mkern-2.3muA^{\star}

Adversarial games

• Adverarial games
• Minimax
• Minimax with \alpha–\beta pruning
• Chess
• Horizon problem
• Expectiminimax

Markov models and Markov decision processes

• Markov models
• Markov chains
• Markov decision processes
• Q learning for MDPs

Decision tree, random forest, and ensemble learning

• Decision trees, introduction
• Mitchell’s canonical example and introduction to ID3
• Shannon entropy and information gain
• Mitchell’s canonical example, continued
• Decision trees: some problems and fixes
• Ensemble learning
• Random forest
• Precision, recall, F1

The neural model and artificial neural networks (forthcoming)