Course Objectives
Statistical Learning refers to a set of tools for modeling and understanding complex datasets. The area combines knowledge from statistics and computer science to tackle these “big data” problems and has become a popular area of work. By end of course, students will be able to formulate prediction problems as statistical machine learning problems, choose and apply the appropriate learning methods to their data, conduct well thought-out statistical machine learning experiments and interpret the results, and write technical reports describing their work.
COVID-19
This will be a learn-as-we-go situation. Please be fexible! We’re going to try to have live lectures on Zoom, which I will record and post to Canvas.
Prerequisites
DSCI 320, DSCI 369 and STAT 341.
Texts
An Introduction to Statistical Learning with Applications in R (2017) by Gareth James, Daniela Witten, Trevor Hastie and Robert Tibshirani – Available free here: http://faculty.marshall.usc.edu/gareth-james/ISL/
Optional Reference: The Elements of Statistical Learning: Data Mining, Inference, and Prediction (2009) by Trevor Hastie , Robert Tibshirani, and Jerome Friedman – Available free here: https://web.stanford.edu/~hastie/ElemStatLearn/
Computing
We will use RStudio (https://rstudio.com), R (https://r-project.org), GitHub (https://github.com), and ggplot2 (https://ggplot2.tidyverse.org). All software is free and open source.
Please install on your own computer or use RStudio Cloud (details to follow).
For your homeworks, you are free to use any other language you like, but I may not be able to help you with your computing if the language is unfamiliar to me.
Classwork and Grading
All graded classwork must be fully reproducible by the instructor and TA. In other words, we need to be able to run your code and have it produce the product you turned in. If this is not the case, it will be reflected in the grading. A copy of your homework will need to be turned in to https://canvas.colostate.edu and the corresponding document used to generate your homework will need to exist on https://rstudio.cloud for full credit.
Homework (70%) Homework will be assigned bi-weekly. All homework assignments are due at 4pm on the due date. Each homework assignment will receive equal weight in the final grade and the one lowest homework assignment grades will be dropped. Late work is not accepted except in rare cases (see Documented emergencies below).
Project (30%) There will be a final project that will consist of an analysis of real data using the tools learned in class (this can include participation in an online data science competition, e.g. Kaggle).
You will write a paper and give an in class presentation. More details will be announced later.
Grades will be assigned according to the following intervals:
| A | A- | B+ | B | B- | C+ | C | D | F |
| [100, 95] | (95, 90] | (90, 87] | (87, 83] | (83, 80] | (80, 77] | (77, 70] | (70, 60] | (60, 0] |
Any grading dispute must be submitted in writing to me within one week after the work is returned.
Extra credit Any extra credit will be announced in lecture only. If you miss lecture, you may miss chances for extra credit.
Policy Regarding Academic Honesty
Statisticians need to have high ethical standards. Thus, I expect each of you to hold high ethical standards and to act with academic integrity in this class. If you have questions about what integrity means, please feel free to ask me. Behavior that will not be tolerated in this class includes turning in a copy of somebody else’s homework or code as your own, copying from somebody’s exam, or failure to cite sources.
This course adheres to the CSU Academic Integrity Policy as found on the Students’ Responsibilities pages of the CSU General Catalog in the Student Conduct Code. Violations will result in zero points for the assignment as a minimum penalty. In addition, CSU policy requires instructors to report violations to CSU’s Office of Conflict Resulution.
Documented Emergencies
If you have a problem that will require you to miss a due date, please discuss this with me in advance if possible. I can grant a rare exception when the reason relates to severe and unavoidable medical or personal emergency. Documentation will be required. Things that typically are not an emergency: vacation, family reunions, ordinary work commitments, job seeking, or other voluntary events. Please schedule these so that they do not conflict with your classes.
Support Services Available
CSU COVID-19 Recovery Page On our road to recovery during these unprecedented times, Colorado State University is committed to the health of our students, faculty and staff, as well as to the health of our university and our ability to continue to empower our community through our land-grant mission of academics, research and outreach.
CSU Health Network Counceling Services A variety of services are offered (151 W. Lake St., Drop-in hours: Monday-Friday 9am-4pm). If you are having difficulty coping, are feeling depressed, or need other psychological assitance, please contact the counceling center.
CSU Disability Center Located in the TILT building. Students with both permanent and temporary limitations and health conditions (physical and mental health) are eligible for support. If you need specific accommodations in this class, please meet with me outside of class to discuss your needs as early in the class as possible.
CSU TILT The Institute for Learning and Teaching has programs to help students improve their study habits, reduce test anxiety, learn about academic integrity, and more.
Disclaimer
I reserve the right to make amendments to the syllabus and the schedule throughout the semester. Any updates will be posted on the class website and announced via e-mail and in class.