Main Page
Statistical and Discrete Methods for Scientific Computing
CSE383M and CS395T, Spring, 2012
Welcome to the course! The instructor is Professor William Press (Bill), and the TA is Jeff Hussmann (Jeff). We meet in CBA 4.344, Mondays and Wednesdays 1:30 - 3:00 pm with Bill, and Fridays, 1:30 - 3:00 p.m. with Jeff. The course is aimed at first or second year graduate students, especially in the CSEM, CS, and ECE programs, but others are welcome. You'll need math at the level of at least 2nd year calculus, plus linear algebra, plus either more continuous math (e.g., CSEM students) or more discrete math (e.g., CS and ECE students).
Problem Sets
These are not to be turned in, but you may be asked in class to discuss your solutions.
Feedback
What has worked well in class so far? What hasn't worked? How could things be improved? Please leave feedback.
Python
Resources for learning Python can be found here.
In-Class Activity Scratch Space
Write Your Own Prelims Exam Mar 19
Sets 24 and 25 Concepts Review Mar 26
Quiz Mar 26 and Histogram of Scores
Term Projects
A term project is required. Please read Term Project Assignment for complete information and calendar of events.
Sign Up Now for Final Oral Exams (Interviews)
Sign up now for your final oral interview (equivalent of final oral exam). The interview will be 20 minutes in length. If you sign up early, you'll get a better time slot! The sign-up page is here.
- Here is the Concepts Study Page and random term generator.
- All oral interviews are in ACES 3.258.
Link to Study Sheet from Last Year
Study Sheet Random term generator
Webcast Lecture Segments (Opinionated Lessons in Statistics)
All of the lectures are in the form of webcasts, divided into segments of about 15-30 minutes each (occasionally a bit longer). Each segment, has a wiki page, page links below. You can view the lecture on its wiki page, which also has additional stuff about the segment, or by clicking directly to YouTube, where they are all on Bill's "Opinionated Lessons" channel. Click on dates below to find files that we used in class, or (sometimes) Bill's or Jeff's solutions to those problems.
watch before Fri Jan 20:
Segment 1. Let's Talk about Probability (or YouTube)
watch before Mon Jan 23:
Segment 2. Bayes (or YouTube)
Segment 3. Monty Hall (or YouTube)
watch before Wed Jan 25:
Segment 4. The Jailer's Tip (or YouTube)
Python walk through on Fri Jan 27
watch before Mon Jan 30:
Segment 5. Bernoulli Trials (or YouTube)
Segment 6. The Towne Family Tree (or YouTube)
watch before Wed Feb 1:
Segment 7. Central Tendency and Moments (or YouTube)
Fri Feb 3
watch before Mon Feb 6:
Segment 8. Some Standard Distributions (or YouTube)
Segment 9. Characteristic Functions (or YouTube)
watch before Wed Feb 8:
Segment 10. The Central Limit Theorem (or YouTube)
Fri Feb 10
watch before Mon Feb 13:
Segment 11. Random Deviates (or YouTube)
Segment 12. P-Value Tests (or YouTube)
watch before Wed Feb 15:
Segment 13. The Yeast Genome (or YouTube)
Fri Feb 17
watch before Mon Feb 20:
Segment 14. Bayesian Criticism of P-Values (or YouTube)
Segment 16. Multiple Hypotheses (or YouTube) Note - the dates attached to segments 15 and 16 have been swapped.
watch before Wed Feb 22:
Segment 15. The Towne Family - Again (or YouTube)
Fri Feb 24
watch before Mon Feb 27:
Segment 17. The Multivariate Normal Distribution (or YouTube)
Segment 18. The Correlation Matrix (or YouTube)
watch before Wed Feb 29:
Segment 19. The Chi Square Statistic (or YouTube)
Fri Mar 2
watch before Mon Mar 5:
Segment 20. Nonlinear Least Squares Fitting (or YouTube)
Segment 21. Marginalize or Condition Uninteresting Fitted Parameters (or YouTube)
watch before Wed Mar 7:
Segment 22. Uncertainty of Derived Parameters (or YouTube)
SPRING BREAK HERE
watch before Wed Mar 21:
Segment 23. Bootstrap Estimation of Uncertainty (or YouTube)
Fri Mar 23
watch before Mon Mar 26:
Segment 24. Goodness of Fit (or YouTube)
Segment 25. Fitting Models to Counts (or YouTube)
watch before Wed Mar 28:
Segment 26. The Poisson Count Pitfall (or YouTube)
watch before Mon Apr 2:
Segment 27. Mixture Models (or YouTube)
Segment 28. Gaussian Mixture Models in 1-D (or YouTube)
watch before Wed Apr 4:
Segment 29. GMMs in N-Dimensions (or YouTube)
Fri Apr 6
watch before Mon Apr 9:
Segment 30. Expectation Maximization (EM) Methods (or YouTube)
Segment 31. A Tale of Model Selection (or YouTube)
watch before Wed Apr 11:
Segment 32. Contingency Tables: A First Look (or YouTube)
Fri Apr 13
watch before Mon Apr 16:
Segment 33. Contingency Table Protocols and Exact Fisher Test (or YouTube)
Segment 34. Permutation Tests (or YouTube)
watch before Wed Apr 18:
Segment 35. Ordinal vs. Nominal Contingency Tables (or YouTube)
Fri Apr 20
watch before Mon Apr 23:
Segment 36. Contingency Tables Have Nuisance Parameters (or YouTube)
Segment 39. MCMC and Gibbs Sampling (or YouTube)
watch before Wed Apr 25:
Segment 40. Markov Chain Monte Carlo, Example 1 (or YouTube)
Segment 41. Markov Chain Monte Carlo, Example 2 (or YouTube)
watch before Mon Apr 30:
Segment 47. Low-Rank Approximation of Data (or YouTube)
watch before Wed May 2:
Segment 48. Principal Component Analysis (PCA) (or YouTube)
Segment 49. Eigenthingies and Main Effects (or YouTube)
Extra Segments (not enough class weeks to assign!)
Segment 37. A Few Bits of Information Theory (or YouTube)
Segment 38. Mutual Information (or YouTube)
Segments with Slides But Not Yet Recorded
(links are to PowerPoint files)
Segment 15.5. Poisson Processes and Order Statistics
Segment 42. Wiener Filtering
Segment 43. The IRE Lady
Segment 44. Wavelets
Segment 45. Laplace Interpolation
Segment 46. Interpolation On Scattered Data
Segment 50. Binary Classifiers
Segment 51. Hierarchical Classification
Segment 52. Dynamic Programming
The Experiment: Active Learning in a Graduate Course?
While the material covered is nearly the same as in previous years, the structure of the course this year is completely different. Much research shows that lecture courses, where students listen passively as the instructor talks, are inefficient ways to learn. What works is so-called active learning, a broad term that, for us, basically means that class time is too valuable to waste on lectures. (See image at right.)
Over the 2011-2012 holiday vacation, Bill converted all the lectures into webcasts and posted them on YouTube under the series title, Opinionated Lessons in Statistics (also see links below). Webcasts are not active learning. However, they will be your main "linear" introduction to the material. You must watch the assigned webcasts before the class for which they are scheduled; maybe watch them more than once if there are parts that you don't easily understand. Then, you will be ready for the active learning that we do in class. The class activities will not "cover the material". Rather, class is supposed to be for "aha moments" and for "fixing" the material in your learning memory.
What are the class activities? Will this work? We will answer those questions by all inventing together. The traditional lecture course is so ... yesterday! What should a mathematics-rich graduate level course for today and tomorrow actually be?
Team Randomizer
Link to the team randomizer
