Artificial Intelligence

Syllabus

1. Introduction (4 hours)
   1. Definition, foundation, history of AI
   2. Importance of knowledge and learning
   3. Cognition and learning (neuroscience)
   4. Human intelligence and machine intelligence
   5. AI tree: branches and interdisciplinary nature
   6. Intelligent agents and types
   7. Agentic AI
2. Problem Solving and Search (9 hours)
   1. Formal problem definition: states, actions, transitions, well-defined problems
   2. Constraint satisfaction problems: node consistency, path consistency, backtracking
   3. Search algorithms, strategies and evaluations
   4. Uninformed search: BFS, DFS, iterative deepening
   5. Informed search: best first search, greedy search, A* algorithm
   6. Adversarial search: minimax algorithm, alpha-beta pruning
   7. Local search and optimization: hill climbing, simulated annealing
   8. Evolutionary optimization: genetic algorithm
3. Knowledge Representation and Probabilistic Reasoning (7 hours)
   1. Knowledge-based agent
   2. Knowledge representation techniques and issues in representation
   3. Propositional and predicate logic
   4. Semantic networks, frames and knowledge graph
   5. Review of Bayes' Theorem and probabilistic reasoning
   6. Fuzzy logic: membership functions, fuzzy inference systems
4. Machine Learning Fundamentals (10 hours)
   1. Foundations and four pillars of machine learning
   2. Review of mathematics for machine learning
   3. Optimization: unconstrained, constrained, convex optimization
   4. Review of machine learning pipeline: model development (generative versus discriminative), learning algorithm, capacity, overfitting and underfitting, hyperparameters and validation sets, estimators, bias and variance, review of MLE, MAP and cross entropy
   5. Supervised learning algorithm: decision tree and support vector machine
   6. Unsupervised learning algorithm: t-SNE
   7. Semi-supervised and reinforcement learning
   8. Model evaluation: energy-based indicators
5. Neural Networks and Deep Learning Algorithms (8 hours)
   1. Neural networks: structures, activation functions and universal approximation theorem
   2. Perceptron, multilayer perceptron and backpropagation
   3. Introduction to deep learning
   4. Concepts on recurrent and generative neural networks
6. AI Applications (5 hours)
   1. Expert systems: characteristics, architecture, development and applications
   2. NLP: level of analysis, challenges, modern approaches and applications
   3. Robotics and computer vision: fundamentals, components and applications
   4. Sustainable AI systems
7. Emerging Trends (2 hours)
   1. Sequence to sequence models
   2. Federated learning
   3. Edge AI
   4. Ethics and AI: responsible AI

Practicals

1. Intelligent agents and problem formulation
2. Uninformed search techniques
3. Informed (heuristic) search techniques
4. Adversarial search and game playing
5. Constraint satisfaction problems (CSP)
6. Evolutionary computation (genetic algorithms)
7. Knowledge representation (logic, semantic networks, frames)
8. Probabilistic and fuzzy reasoning
9. Machine learning pipeline and data preprocessing
10. Supervised and unsupervised learning
11. Neural networks and deep learning basics
12. Mini project, AI applications and ethics