Artificial intelligence (AI) is a research field that studies how to realize the intelligent human behaviors on a computer. The ultimate goal of AI is to make a computer that can learn, plan, and solve problems autonomously. The main purpose of this course is to allow students understanding what the AI is, current theories and applications of AI. The main topics of this course include: Uninformed and informed Search Methods Local Search Methods Game Playing Neural Networks Reasoning under Uncertainty Artificial Intelligence applications in Engineering. Artificial Intelligence applications in Management.
Week 1-2 Uninformed Search Methods – searching a state space graph, problem representation in terms of states, goal test, operators, state-space graph search formulation, expanding a node, frontier list, partial solution path, solution path, search tree, breadth-first search, depth-first search, chronological backtracking, uniform-cost search, iterative-deepening search, bidirectional search, completeness, optimality, admissibility, time and space complexity, detecting repeated states, explored list. Week 3-4 Informed Search Methods – Understand heuristic functions, evaluation functions, best-first search, greedy best-first search, beam search, algorithm A, algorithm A*, admissible heuristic. Week 5-7 Local Search Methods – Local search problem formulation, operators, neighborhood, move set, hill-climbing algorithm, local optima problem, hill-climbing with random restarts, stochastic hill-climbing (simulated annealing) algorithm, escaping local optima, Boltzman’s equation, cooling schedule, genetic algorithms, particle swarm optimization. Week 8-9 Game Playing –game playing as search, search tree, branching factor, minimax algorithm, alpha-beta pruning algorithm, best case and worst case of alpha-beta vs. minimax, iterative-deepening with alpha-beta. Week 9 Mid-term evaluation Week 10-12 Neural Networks –input units, output units, weight space, multi-layer feed-forward networks, hidden units, sigmoid function, back-propagation algorithm, deep learning. Week 13-14 Reasoning under Uncertainty – Random variable, conditional probability, posterior probability, product rule, chain rule, conditionalized version of chain rule, Bayes’s rule, conditionalized version of Bayes’s rule, conditional independence, naïve Bayes classifier, Bayesian networks. Week 15-16 Selected papers discussion and representation of Artificial Intelligence applications in Engineering. Week 17-18 Selected papers discussion and representation of Artificial Intelligence applications in Management.
• Homeworks, paper study and discussion, class participation: 40% • Midterm project: 30% • Final project: 30%
Reference textbook: Artificial Intelligence: A Modern Approach, Russell, Stuart J.; Norvig, Peter