CSE 150 - Summer 2014

Introduction to Artificial Intelligence:
Probabilistic Reasoning and Decision Making


Course calendar | Syllabus | Piazza | GradeSource


A note about the course

This course is essentially the same as my advisor Prof. Lawrence Saul's CSE 150 course. The faster pace of a summer course may make it more challenging --- for both students and instructors! The instructors (lecturer Matt, TA Fred, tutor Raymond) will strive to be as available and responsive as possible. Each will have 2 office hours per week, and we will check Piazza regularly. Please use Piazza! I strongly encourage students to not only ask, but also ANSWER questions on Piazza. One of the best things about college (and grad school!) is learning from your peers. There is no participation grade, but Piazza activity is one factor I will consider if a student's final average falls between grades. Lastly, I have an open door policy: find me in CSE 3242. :-)
-- Matt

Subject

This course will introduce students to the probabilistic and statistical models at the heart of modern artificial intelligence. Specific topics to be covered include: probabilistic methods for reasoning and decision-making under uncertainty; inference and learning in Bayesian networks; prediction and planning in Markov decision processes; applications to intelligent systems, speech and natural language processing, information retrieval, and robotics.

Prerequisites

This course is aimed very broadly at undergraduates in mathematics, science, and engineering. Prerequisites are elementary probability, linear algebra, and calculus, as well as basic programming ability in some high-level language such as MATLAB, R, Python, Java, or C++. (Programming assignments are completed in the language of the student's choice.) Students of all backgrounds are welcome.

Texts

The course will not closely follow a particular text. Lectures are designed to be self-contained. The following optional texts may be useful as general references:

Instructors

Meetings

Grading

Policy

Course calendar

MTuWThF
08/0408/0508/0608/0708/08
Lecture 01Lecture 02
HW 1 out
Lecture 03
Matt OH: 2-4pm
Lecture 04
Ray OH: 1-2pm
Discussion: 5-6pm
HW 1 due, HW 2 out
Fred OH: 12-2pm
08/1108/1208/1308/1408/15
Lecture 05Lecture 06
HW 2 due, HW 3 out
Ray OH: 9-11am
Lecture 07
Matt OH: 2-4pm
Lecture 08
Discussion: 5-6pm
HW 3 due, HW 4 out
Fred OH: 12-2pm
08/1808/1908/2008/2108/22
Lecture 09
Practice midterm out
Lecture 10
HW 4 due, HW 5 out
Ray OH: 9-11am
Lecture 11
Matt OH: 2-4pm
Lecture 12
Discussion: 5-6pm (review for midterm)
Fred OH: 12-2pm
08/2508/2608/2708/2808/29
MIDTERMLecture 13
HW 5 due, HW 6 out
Ray OH: 9-11am
Lecture 14
Matt OH: 2-4pm
Lecture 15
Discussion: 5-6pm
Fred OH: 12-2pm
Practice final out
09/0109/0209/0309/0409/05
NO CLASS
(Labor Day)
Lecture 16
HW 6 due
Ray OH: 9-11am
Lecture 17
Matt OH: 2-4pm
Lecture 18
Discussion: 5-6pm (review for final)
FINAL EXAM: 11:30am-2:30pm, WLH 2207

Syllabus

WEEK 1
MAug 04AI overview, course preview
TuAug 05Modeling uncertainty, review of probabilityHW 1 out
WAug 06Examples of probabilistic reasoning
ThAug 07Belief networks: from probabilities to graphs
FAug 08HW 1 due, HW 2 out
WEEK 2
MAug 11Conditional independence, d-separation
TuAug 12Inference in polytrees and loopy networksHW 2 due, HW 3 out
WAug 13Learning, maximum likelihood estimation
ThAug 14Naive Bayes and Markov models
FAug 15HW 3 due, HW 4 out
WEEK 3
MAug 18Latent variable models, EM algorithmPractice midterm out
TuAug 19Examples of EM algorithmHW 4 due, HW 5 out
WAug 20Hidden Markov models, speech recognition
ThAug 21Viterbi and forward-backward algorithms, belief updating
FAug 22
WEEK 4
MAug 25MIDTERM
TuAug 26Learning in HMMsHW 5 due, HW 6 out
WAug 27Reinforcement learning, Markov decision processes
ThAug 28Policy evaluation, improvement, and iteration
FAug 29Practice final out
WEEK 5
MSept 01NO CLASS (Labor Day)
TuSept 02Value iterationHW 6 due
WSept 03Course wrap-up, review for final exam
ThSept 04Review for final exam
FSept 05FINAL EXAM: 11:30am - 2:30pm, WLH 2207