Course description:

Basic principles and techniques in data analytics; methods for the collection of, storing, accessing, and manipulating standard-size and large datasets; and various techniques of data visualization. The concepts will be implemented with R.

A technical analytics course that will teach supervised and unsupervised learning strategies, including regression, generalized linear models, regularization, dimension reduction methods, tree-based methods for classification, and clustering. Upper-level analytical methods are shown in practice: e.g., neural networks and Gaussian processes.

Prerequisites:  CMDA/CS/STAT-3654, and STAT-3104 or STAT-4705 or STAT-4714 or CMDA-2006

Students must be familiar with intermediate linear algebra, calculus, probability, basic statistics (standard errors), regression, and coding in a high-level language R/Python/Matlab. (Basically, you need mathematical and computational maturity, and this can’t be your first course in Stats!)

Required Course Materials:

Required Text: There is no required text for this course.

Optional Texts:

●   An Introduction to Statistical Learning with Applications in R (ISLR), Gareth James, Daniela Witten, Trevor Hastie and Robert Tibshirani. Free pdf here .

●   Pattern Recognition and Machine Learning, Cristopher M. Bishop. Google the title and you will find a free pdf.

● Machine Learning: a Probabilistic Perspective, Kevin Murphy.

Software:

We will be using statistical software in this class. You are welcome to use the software of your choice, but class demonstrations/notes will be in R only. All help with software in office hours will be limited to R. Please install R and RStudio as soon as possible. R code supporting the ISLR book can be found on  the  book’s  webpage.  Matlab  code  supporting  the  Bishop book  can be  downloaded  from http://prml.github.io/.

Assessments and Grade Weighting:

Your grade will be determined based upon the following:

● Homework/In-Class Assignments: 50%

● Projects/Exams: 50%

The grading scale will be as follows:

●   [ 92.5, ∞ )       A

●    [ 89.5, 92.5 )    A-

●    [ 86.5, 89.5 )     B+

●    [ 82.5, 86.5 )    B

●    [ 79.5, 82.5 )     B-

●    [ 76.5, 79.5 )    C+

●   [ 72.5, 76.5 )    C

●    [ 69.5, 72.5 )    C-

●    [ 66.5, 69.5 )    D+

●    [ 62.5, 66.5 )    D

●    [ 59.5, 62.5 )    D-

● [ 0, 59.5 )         F

Attendance Policy:

Attendance is strongly encouraged. If you are feeling ill, especially with COVID-19 or flu-like symptoms, please don’t come to class.

We will be following Virginia Tech’s updated guidelines regarding sickness. This means that we will not simply take your word that you are ill. You need to obtain a note from a doctor’s office or health clinic. You can then submit your note to the Dean’s office who will contact all of your instructors.

Submitting Completed Work on Canvas:

All completed assignments/exams/projects will be submitted to Canvas only. Do not email the professor or GTAs your assignment materials to be graded. If the assignment is not on Canvas, it will not be graded.

Extensions will not be granted under any circumstance, beyond the extenuating circumstances that have been cleared by the Dean’s office.

Late assignments will not be accepted. If an assignment is due at a specific time on Canvas, that does not mean wait until the last minute to turn it in. For example, a deadline of 1pm does not mean wait until 1pm then turn it in. Turn it in BEFORE 1pm. We will not accept excuses that you “finished the assignment, but forgot to turn in the assignment, here is my screenshot to prove it” and other similar statements. Students have doctored such images in the past and have been reported for honor code violations. You will receive ample time to complete all assignments for this course and aside from illness related issues, you have no excuse from getting the material turned in before the deadlines. It is your responsibility to double check the submission to make sure that it is correctly uploaded and is the correct file.

CMDA Statement on Academic Integrity:

Please see the CMDA Statement on Academic Integrity provided on the Canvas page. You are expected to adhere to the policies outlined in this document which also provides some examples of violations that will result in a student potentially being reported for an honor code violation.

In general, students are encouraged to discuss homework assignments in groups unless otherwise specified by the professor. However, the student’s submitted work should represent their own ideas.

Students are not allowed to copy computer codes or answers from each other, and must write their own codes and solutions.

Work Presentation:

Your work should be professional quality. A combination of R Markdown and/or LaTeX and related tools is required, unless otherwise specifically stated. We will not grade raw computer code that is not contained in a compiled report, this includes raw R Markdown and/or LaTeX code sent in .R/.Rmd/.tex files. Failure to compile your work properly using R Markdown and/or LaTeX will result in a zero as this is an essential base skill.  Additionally, you must compile your work directly to a .pdf file (conversions from Word/HTML files are not acceptable).

In some rare instances, hand-written work will be allowed but this too must be scanned in and embedded in an R Markdown document.

The grader reserves the right to mark off for untidy or unclear submitted work.  Each problem must be clearly labeled and final answers clearly indicated. Problems must be done in the correct order, and your name must appear on the first page. You must show all work for consideration of partial credit.

Homework & In-Class Assignments:

Homework will be assigned and due on a regular basis. Homework will consist of both traditional mathematical, short-answer type questions, and computer programming. All work must be shown for credit and you must provide your computer code, when appropriate, with attached software output.

Each HW will be worth 100 points, unless otherwise stated.

Solutions to homework assignments will not be provided. The TAs will be providing you with feedback as to what you got incorrect in the traditional manner. If you need extra clarity on why you missed something, please discuss this with them directly.

We will also be doing in-class work on a semi-regular basis.  These exercises may be stand-alone (separate from the regular homework) assignments or may be combined with a larger homework assignment. You are expected to know whether such assignments have been given during the class period and be aware of their due dates as these might have shorter deadlines than regular homework (e.g. by the end of class or midnight on the same night).

Projects & Exams:

I will use a mixture of projects and/or exams during this course. Depending on how I feel the flow of the timeline for the course is progressing, I will either have a mixture of exams/projects or I may do entirely either just exams or just projects.

Exams may be either in-class or take-home. Exams will be announced no less than one week prior to the exam date.

Projects may either be assigned to individuals, pairs of students, or to small groups. Projects will typically consist of a combination of concepts and application of methods. Starting projects at the last minute is a recipe for disaster, so begin early!

We will make use of peer reviews for providing feedback and assisting in the grading of the projects. However the professor will be assigning the final grade on all projects. Rubrics will be provided. Additional instructions will be provided when the project is assigned. Failure to grade within guidelines will impact your own grade on the project as a component of grade is reserved for participation in the peer review.

Other class policies:

●   Barring illness or other unforeseen emergencies, missed homework assignments and projects cannot be made up and will receive a grade of zero.

●   Any questions about homework grades should be referred first to the graduate teaching assistant using the comment box on Canvas. If the TA does not respond via Canvas within

24 hours, then you can email them directly. If you still have questions about a grade, or cannot resolve an issue with the teaching assistant, then contact the professor.

●   The tenets of the Virginia Tech Honor Code will be strictly enforced in this course and all assignments shall be subject to the stipulations of the Honor Code.