Content
1. Quantifying uncertainty: acting under uncertainty, inference using full joint distributions, variable independence, Bayes rule, probabilistic puzzles
2. Probabilistic reasoning I: Bayesian networks (BNs), global and local semantics, constructing BNs, Markov blanket, d-separation algorithm, deterministic nodes, noisy-or;
3. Probabilistic reasoning II: exact inference (enumeration, variable elimination), approximate inference (rejection sampling, likelihood weighting), markov chains;
4. Probabilistic reasoning over time I: time and uncertainty, inference in temporal models;
5. Probabilistic reasoning over time II:hidden Markov models, Kalman filters, dynamic Bayesian networks, Viterbi algorithm;
6. Making simple decisions: utility theory, utility functions, decision networks, the value of information, cognitive biases;
7. Making complex decisions: sequential decision problems, value iteration, policy iteration, partially observable Markov Decision Processes, game theory, mechanism design;
8. Supervised learning: regression and classification, decision trees, regression trees, learning probabilistic models, Naive Bayes, learning with hidden variable, support vector machines, ensemble learning
9. Artificial neural networks: perceptrons, multilayer perceptrons, backpropagation, deep learning (DL), convolutional neural networks, recurrent neural networks; applications of DL
10. Knowledge in learning: explanation-based learning, learning using relevance information, inductive logic programming, version spaces, explainable AI (XAI)
11. Unsupervised learning: datamining, cluster analysis, partitional clustering, k-means, bisecting k-means, hierarchical clustering, cluster similarity;
12. Unsupervised learning: frequent itemset generation, rule generation, compact representation of frequent itemsets, sequential pattern mining
13. Reinforcement learning (RL): passive RL, active RL, policy search, applications of RL;
14. Natural language processing: machine comprehension, augmented grammars and semantic representation, text classification, AI ethics.