SYDE 522 – Machine Intelligence (Winter 2018, University of Waterloo)
Target Audience: Senior Undergraduate Engineering Students
Instructor: Professor H.R.Tizhoosh (http://kimia.uwaterloo.ca/)
Course Outline - The objective of this course is to introduce the students to the main concepts of machine intelligence as parts of a broader framework of “artificial intelligence”. An overview of different learning, inference and optimization schemes will be provided, including Principal Component Analysis, Support Vector Machines, Self-Organizing Maps, Decision Trees, Backpropagation Networks, Autoencoders, Convolutional Networks, Fuzzy Inferencing, Bayesian Inferencing, Evolutionary algorithms, and Ant Colonies.
Lecture 1 – Introduction (Definition of Intelligence, terminology, history of AI, Turing Test, Chinese Room)
Lecture 2 – Principal Components Analysis (PCA)
Lecture 3 – Linear Discriminant Analysis (LDA), t-distributed Stochastic Neighbor Embeddings (t-SNE)
Lecture 4 – AI and Vision, feature extraction, Harris corners, Fisher Vector, VLAD, SIFT, Bag of Visual Words
Lecture 5 – AI and data, generalization and memorization, K-fold cross-validation, leave-one-out, regularization, overfitting, underfitting
Lecture 6 – Clustering, K-means, self-organizing maps (SOM)
Lecture 7 – Classification, support vector machines (SVM)
Lecture 8 – Cluster validity, SSW, SSB, Dunn’s Index, WB Index, Fuzzy sets and fuzzy c-means (FCM)
Lecture 9 – Linear regression, artificial neurons, abstractions of neurons, weight adjustment for plasticity
Lecture 10 – Artificial Neural Networks, XOR problem, hidden layers, learning algorithm, multi-layer perceptrons (MLPs), Delta Rule
Lecture 11 – Backpropagation networks (incremental and batch-wise), stopping criteria, autoencoders
Lecture 12 – Restricted Boltzmann Machines (RBMs), training deep autoencoders
Lecture 13 – Neocognitron, Convolutional Neural Networks (CNNs), overfitting in deep learning
Lecture 14 – Reinforcement Learning, reward and punishment
Lecture 15 – Designing Reinforcement Learning agents, Temporal differencing, Q-learning
Lecture 16 – Decision Trees, entropy, information gain
Lecture 17 – Fuzzy Logic, modus ponens, modus tollens, inference, fuzzy control, inverted pendulum
Lecture 18 – Bayesian Learning, Bayes Theorem, probability rules
Lecture 19 - Naïve Bayes classifier
Lecture 20 – Evolutionary algorithms, genetic algorithms
Lecture 21 – Genetic algorithms: encoding, crossover and mutation, different models, differential evolution, opposition-based learning
Lecture 22 – Swarm Intelligence, Ant Colony Optimization (ACO)
Lecture 23 – Ethics of Artificial Intelligence, Part 1 (Philosophy)
Lecture 24 – Ethics of Artificial Intelligence, Part 2 (Practical Cases)