University of Alaska Fairbanks MAYMester: May 6 - 17, 2019
Sign up details here! https://www.uaf.edu/summer/sessions/maymester/ https://www.alaska.edu/coursefinder/framework/index.xml?CRN=51833&term=201902
This is an intensive 10 day course on the theory and practice of Machine Learning as applied to environmental sciences such as fisheries, oceanography, environmental chemistry, geography, hydrology, wildlife biology, ecology, anthropology, demography, etc. Recent advances in computer power have enabled data scientists to analyze ever-larger datasets using machine learning and artificial intelligence techniques, making the ability to process these datasets an integral training requirement for the next generation of environmental scientists. Through hands on programming, this course will focus on training students in environmental fields in the use of modern computational tools for machine learning applications. Students will work on group projects and may bring their own data to use as examples in homework exercises.
Out Of State and International Students Welcome! Everyone pays in-state tuition for summer courses, and housing is available on campus.
https://www.uaf.edu/summer/whoareyou/new/outofstate/
https://www.uaf.edu/summer/whoareyou/new/international/
Instructor: Eric Collins (recollins alaska edu)
Meeting time: M-F 9:00am – 1:00pm
Textbook: None. Readings will be taken from the primary literature.
Supplemental Reading: “Applied Predictive Modeling” by Kuhn and Johnson. ISBN: 9781461468486 (recommended)
Supplies: Internet-enabled portable computer is required. A dozen laptops are available from the CFOS academic office, please contact the instructor for assistance.
Expected Proficiencies: Undergraduate-level understanding of basic statistics and programming in at least one language (preferably Python, R, or MATLAB).
Prerequisites: Introductory course in programming (CS103, CS201, CS405) or computational data analysis (BIOL680, GEOS436, GEOS636, MSL464, MSL604, MSL627, MSL631, MSL632) or statistics (STAT200, STAT401, STAT641) or graduate standing or by permission of the instructor. (3+0)
Course Goals: The goal of this course is to introduce modern computational tools in machine learning to students in environmental fields like fisheries, oceanography, geography, hydrology, and wildlife biology.
Student Learning Outcomes -- Upon completion of the course students will be able to:
- Use a command-line environment to conduct routine tasks on the computer (e.g. the bash shell).
- Write simple scripts using at least one machine learning toolkit (e.g. scikit-learn or TensorFlow).
- Find, download, install, and use software and datasets from public repositories (e.g. from Github or Google Earth Engine).
- Apply appropriate concepts and algorithms in machine learning to real world problems (e.g. supervised learning, classification).
- Analyze a real environmental sciences dataset using machine learning.
Instructional Methods: The in-class course time will consist of lectures on concepts and algorithms, in addition to a group project where students will gain hands-on experience working with real datasets.
Very Tentative Course Calendar:
Date Topic Homework
Day 1 Introduction and Basic Concepts HW1 assigned: Intro to Python and NumPy
Day 2 Linear and Logistic Regression HW1 due
Day 3 Supervised Learning HW2 assigned: Intro to SciPy and matplotlib
Day 4 Clustering and Dimensional Reduction HW2 due
Day 5 Unsupervised Learning HW3 assigned: Intro to Scikit-learn
Day 6 Data Selection and Factoring HW3 due
Day 7 Model Selection and Optimization HW4 assigned: Intro to TensorFlow
Day 8 Case Study: Image Classification HW4 due
Day 9 Case Study: Prediction
Day 10 Conclusions and Group Presentations
Evaluation: Students will be evaluated based on class participation, homework, a group project, and a final presentation. Grading is absolute.
Class Participation (10%, 100 points), including on-time attendance at lecture and engagement with classmates, will be expected of each student, and evaluated according to the following rubric (also available at https://www.cmu.edu/teaching/assessment/examples/courselevel-bycollege/cfa/tools/participationrubric-cfa.pdf)
Homework in the form of Computational Exercises (40%, 400 points) will be required showing proficiency in the use of machine learning tools. Homework will be provided during class and will be due the next day. Late homework will not be accepted.
A Group Project (40%, 400 points) will be required. The product will be organized as a scientific manuscript (>2500 words) based upon an independent computational analysis using skills developed in class. The topic of the Group Project will be agreed upon with the instructor by Day 4, and may use public datasets or the student’s own dataset. Publishing criteria from the journal PLoS ONE will be used as guidance: http://journals.plos.org/plosone/s/criteria-for-publication
The Final Presentation (10%, 100 points) will be based on the Group Project; students are expected to explain their findings in a professional manner in a 15-minute conference-style presentation on the last day. Course Policies: Students are expected to read the relevant material prior to the lectures and attend class in a timely manner. Active participation is expected. The use of cell phones or other electronic communications (e.g. email, twitter, facebook etc.) during class is considered inappropriate. Students should be familiar with the UAF Honor Code (https://www.uaf.edu/catalog/catalog_00-01/undergrad/regs3.html). Cheating and plagiarism will not be tolerated. Any student found cheating during the exams or to have plagiarized or fabricated statements (including passages from web pages) will receive an automatic 'F' for the class.
The following non-curved grading system will be used for the entire course:
A+ >95%
A >90 – 95%
A− >85 – 90%
B+ >80 – 85%
B >75 – 80%
B− >70 – 75%
C+ >67 – 70%
C >63 – 67%
C− >60 – 63%
Grades below C− will not count toward the major or minor degree requirements
D 50 – 60%
F <50
Support Services: At UAF, the Office of Disability Services (203 Whitaker Bldg; 474-5655; TTY 474-5655; [email protected], http://www.uaf.edu/disability) ensures that students with physical or learning disabilities have equal access to the campus and course materials. If you have specialized needs, please contact this office or the instructor to make arrangements. The UAF Writing Center (801 Gruening Bldg) is available for helping students in brainstorming and generating topics, organizing ideas, developing research strategies, the use of citations, and editing for clarity and correctness. Contact them at http://www.uaf.edu/english/writing-center. Student protections and services statement: Student protections and services statement: Every qualified student is welcome in my classroom. As needed, I am happy to work with you, disability services, veterans' services, rural student services, etc to find reasonable accommodations. Students at this university are protected against sexual harassment and discrimination (Title IX), and minors have additional protections. For more information on your rights as a student and the resources available to you to resolve problems, please go the following site: www.uaf.edu/handbook/