Syllabus: Statistical Modeling with Stochastic Processes.

Description: Stochastic processes are powerful tools for constructing the rich models needed to capture the complexity of our world. We will discuss an array of concrete examples where stochastic processes are used to perform sophisticated statistical inferences. After going through these examples, you will be familiar with the main building blocks, you will know how to compose them to create new models, you will be able to design inference algorithms for your models and you will have a better understanding of the limits of these models and algorithms. Throughout the course there will be an emphasis on computational statistics and on applications in machine learning, phylogenetics, computational biology and linguistics.

Prerequisite: STAT 460 / 560 or MATH 419 or CS 540 or equivalent (if you are not sure, come talk to me after one or two lectures).

Important notes:

• I will try my best to post lecture notes (or at least, draft of) in advance. You can access them in the Lecture Notes tab.
• Attendance at labs optional, but strongly recommended. Some labs may have suggested pre-reading/pre-coding, see the tab Labs above.
• Discussion forum: To encourage richer open exchanges, Seong and I will only use this platform to answer course-related questions (unless for personal matters). See Contact link in the top menu.
• Languages used in the exercises: Java, Bugs, R.
  • R is great to use existing statistical methodologies, but to develop new ones, it is ideal to have other tools in the toolbox.
  • We will organize remedial tutorials for Java/Bugs as we go along. Please try to drop by if you are not familiar with the covered topic.
• Environments supported: Stat net, Mac OS X, linux.

Evaluation

• Weekly exercises (they will involve coding) 45%
• Final research project (teams encouraged) 45%
• Participation: 10%
  • in class and/or labs and/or piazza and/or office hours, reporting typos in notes,
  • scribing/editing activity.

Handing-in exercises:

• Organize your files into zip archives as explained in Exercise 1
• Go to the following URL.
• Enter password bayes (no capitals)
• You will see an upload button after login

Editing duties: Everyone should claim editorship of one lecture. There should be one or two editors per lecture. The editor(s) are responsible for:

1. Adding some supplementary references, notes, observations, etc.
2. Correcting errors and typos.
3. Participating in piazza discussion related to the lecture.
4. Adding the diagrams and figures that are not already included.

The editors should complete these tasks within 1 week of the lecture. This will be coordinated via github. Please create an account if you do not have one already. See below for details.

The editing process:

• On the week that you will be scribing, on the day of the lecture (or before) send in a email to Seong your github account, he will give you access to the website github, available at this address.
• The files corresponding to the lecture notes are available in _posts/. If you have not used git before, you can edit the file directly from the github website (click on a file, and then on the Edit button), but we also encourage you to learn using git, either with a graphical interface (SourceTree or Github's), or via command line (many available via google, for example there).
• Please try as much as possible to edit only the file corresponding your lecture to avoid too many merge conflicts.
• Note also that some changes might be pushed automatically to the website, so make sure to push commit to server only if you are sure the change is good (especially relevant if you are using the git server directly, and this is why we strongly recommend to use your own local git and then push from it).

Final project:

The course project involves independent work on a topic of your choice, with the constraint that you should make use of some of the theory covered in class, or extension of these techniques. There are three main types of projects: application, methodology, and theory, as described in class. Combinations of these is also encouraged. Extending the exercises is a good way to start thinking about project ideas.

• Teams are encouraged, in which case you should outline the final report who did what.
• Format: A latex document (and code if there is an empirical aspect), submitted electronically using the usual method.
• Grading: I will base the grade on the same factors one would usually consider in a paper reviewing process (but taking into account the fact that the time frame is shorter than the typical time to write a research paper). Is the goal clearly defined? Is it well motivated? Is the approach sound? Creative? Is the paper well-written? Are there interesting connections made to the existing literature?

Textbook: There are no textbook. Resources and pointers will be posted in the online lecture notes.

Office hours: We will reserve time at the lab for Q&A. More office hours will be added as needed.

Acknowledgement: computing supported by an AWS in Education Grant award.