CSCE 582 Spring 2020 Time Allocation Framework: Second Continued Operations Revised Syllabus Spring 2020

The following changes were carried out as a consequence of the cancellation of classes during the week of March 17 and the continuation of the course as online only from March 23:

There will be no in-class graduate student presentations, which were originally planned for the last week of classes. Graduate students will present by video link outside of class time, if convenient, or replace the oral presentation with commented slides (PowerPoint or pdf), or with a report.
In-class lectures will be replaced by previously recorded videos. The instructor will be available for questions, answers, reviews of homework, a discussion of Ch.4 [J96], etc., virtually in Blackboard Collaborate Ultra.
The grading policy is unchanged.. (See https://cse.sc.edu/~mgv/csce582sp20/grading.html).
The in-class final exam will be replaced by a take-home final exam.
Other course information is unchanged, except for the cancellation of in-person office hours. (See https://cse.sc.edu/~mgv/csce582sp20/index.html).

The textbook is:

Jensen, Finn V. and Thomas D. Nielsen. Bayesian Networks and Decision Graphs. New York: Springer-Verlag, 2007 (ISBN-10 0-387-68281-3; ISBN-13: 978-0-387-68281-5) (Approximately $70 at amazon.com. This text is referred to as [J07].)

Other recommended books on the topic are:

Adnan Darwiche. Modeling and Reasoning with Bayesian Networks Cambridge, UK: Cambridge University Press, 2009 (ISBN 978-0-521-88438-9).
Daphne Koller and Nir Friedman. Probabilistic Graphical Models: Principles and Techniques. Cambridge, MA: The MIT Press, 2009 (ISBN 978-0-262-0139-2).
Timo Koski adn John M. Noble. Bayesian Networks: An Introduction. Chichester, UK: Wiley, 2009 (ISBN 978-0-470-74304-1).
Uffe B. Kjaerulff and Anders L. Madsen. Bayesian Networks and Influence Diagrams: A Guide to Construction and Analysis. New York: Springer-Verlag, 2008 (ISBN 978-0-387-74100-0).
Kevin B. Korb and Ann E. Nicholson. Bayesian Artificial Intelligence, 2nd ed. London: Chapman & Hall and CRC Press, 2011 (ISBN 978-1-4398-1591-5). (First edition: London: Chapman & Hall and CRC Press, 2004, ISBN 1-58488-387-1.)
Jensen, Finn V. Bayesian Networks and Decision Graphs. New York: Springer-Verlag, 2001 (ISBN 0-387-95259). (This text is referred to as [J01].)
Robert G. Cowell, A. Philip Dawid. Steffen L. Lauritzen, and David J. Spiegelhalter. Probabilistic Networks and Expert Systems. New York: Springer, 1999 (ISBN 0-387-98767-3).
Jensen, Finn V. An Introduction to Bayesian Networks. New York: Springer-Verlag, 1996 (ISBN 0-387-91502-8). (This text is referred to as [J96]. There was a printing of this book in 1995, so it is sometimes dated 1995 instead.)
Richard E. Neapolitan. Probabilistic Reasoning in Expert Systems: Theory and Algorithms. New York: Wiley, 1990 (ISBN 0-471-61840-3).
Judea Pearl. Probabilistic Reasoning in Intelligent Systems. San Francisco: Morgan-Kaufmann, 1988 (ISBN 0-934613-73-7).

The main (approximately 80% of the time) instructional delivery strategy for this course is lectures. Discussions based on student presentations, videos, quizzes or in-class exercises, and a possible invited talk will make up the remaining 20% of time. The first day of classes is Tuesday, January 14, 2020. The last day to drop the course without a grade of "W" being recorded is Tuesday, January 21, 2020. The last day to withdraw without failure is Monday, April 6, 2020. The last day of classes is Thursday, April 23, 2020. Reading Day is Tuesday, April 28, 2020. The final exam for the course is Thursday, April 30, 2020, at 1230 in the classroom (SWGR 2A22). This is the regularly scheduled time for courses taught from 1140 to 1255 on Tuesdays and Thursdays. (See the university exam schedule.) The syllabus is tentative. It is especially important for students to be aware that the date of the midterm exam is tentative.

Week Lecture Topics Readings Homework

1: January 13 and 15 Probability and Reasoning Chs.1 and 2 [J07]; instructor's slides See Lecture Log

2: January 21 and 23 Probability and Reasoning Chs.1 and 2 [J07]; instructor's slides See Lecture Log

3: January 28 and 30 Causal and Bayesian Networks Ch.2 [J07] See Lecture Log

4: February 4 and 6 Causal and Bayesian Networks Ch.2 [J07] See Lecture Log

5: February 11 and 13 Building Models: Capturing the Structure and Determining the Conditional Probabilities Sections 3.1 and 3.2 [J07] See Lecture Log

6: February 18 and 20 Building Models: Capturing the Structure and Determining the Conditional Probabilities Sections 3.2 and 3.3 [J07] and Notes on the Stratum Method See Lecture Log

7: February 25 and 27 Building Models: Advanced Modeling Methods and Special Features Sections 3.3 and 3.4 [J07]

8: March 3 and 5 Midterm and Building Models: Advanced Modeling Methods and Special Features Sections 3.3 and 3.4 [J07] See Lecture Log

9: March 10 and 12 Spring Break No Assigned Readings No Homework Due

10: March 17 and 19 Classes Canceled No Assigned Readings No Homework Due

11: March 24 and 26 Interventions: excision (cut) semantics; forcing variables; example form Korb and Nicholson. Object-oriented BNs. Repetitive models: temporal BNs, Dynamic Bayesian Networks (DBNs), Hidden Markov Models (HMMs), Kalman filters. Sections 3.2.6, 3.3.9 causality and intervention), 3.3.6 (OOBNs), 3.3.7 (DBNs, HMMs, and other repetitive models) [J07]. Video presentations of February 25 (Wed) 2009 and February 27 (Fri), 2009 HW8: Exercises 3.10 (i only), 3.12 (i and ii only), 3.13 (i-iii only)

12: March 31 and April 2 Belief Updating in Bayesian Networks: The Junction Tree Method Ch.4 [J96] (lecture) and Ch. 4 [J07] (offline reading). Video Presentations of March 6 (end), March 16, March 18, March 20, March 23, March 27, March 30 (beginning), all 2009. No Homework Due

13: April 7 and 9 Belief Updating in Bayesian Networks: Stochastic Simulation and Loopy Belief Propagation Sections 4.7-4.8 [J07]; Section 4.6 [J96]. Video presentations of March 30 2009 and April 1 2009. HW9: Exercises 3.16 (use a Bayesian network), 3.19 (Do not use the Noisy-OR approximation and enter your model in Hugin), 3.20, 3.21 (part ii only) (in 3.19, you did not use the Noisy-OR approximation; in 3.21 you use it in the "additive" form of Figure 3.31)

14: April 14 and 16 Graphical Languages for Decision Problems Ch.9 [J07]. Video presentations of April 3 2009 through April 13 2009. HW10: Exercises 3.7 and 3.29

15: April 21 and 23 Graphical Languages for Decision Problems Ch.9 [J07]. Video presentations of April 3 2009 through April 13 2009. No Homework Due

Week	Lecture Topics	Readings	Homework
1: January 13 and 15	Probability and Reasoning	Chs.1 and 2 [J07]; instructor's slides	See Lecture Log
2: January 21 and 23	Probability and Reasoning	Chs.1 and 2 [J07]; instructor's slides	See Lecture Log
3: January 28 and 30	Causal and Bayesian Networks	Ch.2 [J07]	See Lecture Log
4: February 4 and 6	Causal and Bayesian Networks	Ch.2 [J07]	See Lecture Log
5: February 11 and 13	Building Models: Capturing the Structure and Determining the Conditional Probabilities	Sections 3.1 and 3.2 [J07]	See Lecture Log
6: February 18 and 20	Building Models: Capturing the Structure and Determining the Conditional Probabilities	Sections 3.2 and 3.3 [J07] and Notes on the Stratum Method	See Lecture Log
7: February 25 and 27	Building Models: Advanced Modeling Methods and Special Features	Sections 3.3 and 3.4 [J07]
8: March 3 and 5	Midterm and Building Models: Advanced Modeling Methods and Special Features	Sections 3.3 and 3.4 [J07]	See Lecture Log
9: March 10 and 12	Spring Break	No Assigned Readings	No Homework Due
10: March 17 and 19	Classes Canceled	No Assigned Readings	No Homework Due
11: March 24 and 26	Interventions: excision (cut) semantics; forcing variables; example form Korb and Nicholson. Object-oriented BNs. Repetitive models: temporal BNs, Dynamic Bayesian Networks (DBNs), Hidden Markov Models (HMMs), Kalman filters.	Sections 3.2.6, 3.3.9 causality and intervention), 3.3.6 (OOBNs), 3.3.7 (DBNs, HMMs, and other repetitive models) [J07]. Video presentations of February 25 (Wed) 2009 and February 27 (Fri), 2009	HW8: Exercises 3.10 (i only), 3.12 (i and ii only), 3.13 (i-iii only)
12: March 31 and April 2	Belief Updating in Bayesian Networks: The Junction Tree Method	Ch.4 [J96] (lecture) and Ch. 4 [J07] (offline reading). Video Presentations of March 6 (end), March 16, March 18, March 20, March 23, March 27, March 30 (beginning), all 2009.	No Homework Due
13: April 7 and 9	Belief Updating in Bayesian Networks: Stochastic Simulation and Loopy Belief Propagation	Sections 4.7-4.8 [J07]; Section 4.6 [J96]. Video presentations of March 30 2009 and April 1 2009.	HW9: Exercises 3.16 (use a Bayesian network), 3.19 (Do not use the Noisy-OR approximation and enter your model in Hugin), 3.20, 3.21 (part ii only) (in 3.19, you did not use the Noisy-OR approximation; in 3.21 you use it in the "additive" form of Figure 3.31)
14: April 14 and 16	Graphical Languages for Decision Problems	Ch.9 [J07]. Video presentations of April 3 2009 through April 13 2009.	HW10: Exercises 3.7 and 3.29
15: April 21 and 23	Graphical Languages for Decision Problems	Ch.9 [J07]. Video presentations of April 3 2009 through April 13 2009.	No Homework Due