Artificial Intelligence - ECSE 526
course description
This course will introduce students to Artificial Intelligence (AI), beginning from historical and philosophical perspectives, progressing through a number of core topics from classical AI, and then dealing extensively with various areas of machine learning. The latter topic will emphasize connectionist architectures (artificial neural networks) and evolutionary computing approaches.
What will I learn?
Is this course for me?
What's the workload like?
What's the text book?
How is my grade determined?
What are the in-class and homework activities?
What are the due dates and policy on late work?
Where's that line about academic integrity?
(Subject to minor revision throughout the term)
Date Topic and slides Readings Videos Material Due weight
Sep. 2 Introduction: administrivia, definitions, historical and philosophical underpinnings, and what this course is about Ch 1.1, 1.3-1.3.5 | Littman, 'Rise of the Machines' is Not a Likely Future | Social Robots  
Sep. 7 Problem Solving By Search: problem formulation, search strategies, heuristics Ch. 3 |    
Sep. 12 Two-player games: minimax, alpha-beta pruning, position evaluators Ch. 5-5.4 | Patrick Winston. minimax lecture | Francisco Iacobelli. minimax tutorial  
Sep. 14 Intelligence and agency: agent classifications and architectures Ch. 2 | Brooks, A Robust Layered Control System for a Mobile Robot |  
Sep. 19 Constraint Satisfaction Problems Ch. 6-6.4    
Sep. 21 Logical Agents: wumpus world, reasoning in propositional logic Ch. 7-7.5    
Sep. 26 Planning Ch. 10-10.2 | Dolphin planning  
Sep. 28 in-class game tournament Assignment #1 15%
Oct. 3 Probability Basics Ch. 13.2-13.5 Jeff Miller, part 1 | part 2 | Patrick Winston, lecture | Jeff Miller, tutorial on proportionality | Nando de Freitas, Bayes' Rule  
Oct. 5 Hidden Markov Models Ch. 15-15.3, Eisner lecture and interactive spreadsheet Jeff Miller, part 1 | part 2  
Oct. 12 Decision Making: simple and complex decisions, Markov decision problems, utility, value iteration, policy iteration Ch. 16.3, 17-17.4    
Oct. 17 Principles of Machine Learning: regression, overfitting, training, validation, and test sets, leave-one-out cross-validation, k-fold cross-validation, gradient descent Ch 18-18.2 | Domingo, A Few Useful Things to Know about Machine Learning Jeff Miller, supervised learning | unsupervised learning  
Oct. 19 Inductive Learning: decision trees, pruning, ensemble learning, boosting Ch. 18.3 | Sammut et al., Learning To Fly Jeff Miller, decision trees Assignment #2 15%
Oct. 24 Passive Reinforcement Learning: direct utility estimation, adaptive dynamic programming, temporal difference learning Ch. 21-21.2 |    
Oct. 26 Active Reinforcement Learning: action-value function, Q-TD, exploratory learning, generalization Ch. 21.3-21.6 |   exercise
Oct. 31 Non-parametric models: nearest neighbour, kd-trees, locally sensitive hashing Ch. 18.8 | Mitchell, Instance-based learning (pp. 230-234) Patrick Winston, lecture  
Nov. 2 k-means and PCA: clustering, component analysis, and dimensionality reduction PCA and Self-organizing maps | Turk, Face recognition using eigenfaces Jeff Miller, short tutorial  
Nov. 7 Statistical Learning: Bayesian learning, MAP, maximum likelihood Ch. 20-20.2 Jeff Miller short tutorial  
Nov. 9 Mixture Models and Expectation Maximization Ch. 20.3 | Jeff Miller, EM algorithm | Why EM makes sense, part 1 | part 2  
Nov. 14 in-class game tournament Assignment #3 15%
Nov. 16 Intro to Artificial Neural Networks Ch. 18.7 | Mitchell, Artificial Neural Networks (4.1-4.6) |
Schraudolph & Cummins, Introduction to NN | Tveter, BackProp Basics (or available as PDF here | Can animals deduce? (video) | Rowley, Baluja & Kanade, Neural-Network-Based Face Detection | Fels, An Adaptive Interface that Maps Hand Gestures to Speech |
Welch Labs, playlist  
Nov. 21 Neural Network Applications OCHRE applet |   exercise
Nov. 23 Evolutionary Computing: genetic algorithms, genetic programming, evolving neural networks Nolfi et al Phenotypic Plasticity in Evolving NN | Bling, MarI/O (video) | Koza, Breeding Populations of Programs to Solve Problems in AI |    
Nov. 28 Project Presentations  
Nov. 30 Project Presentations  
Dec. 5 (no class) Project 25%
Last updated on 9 December 2016