Course Information
Instructor: Karianne Bergen, Assistant Professor of Data Science and EEPS and Assistant Professor of Computer Science
Faculty Office hours: by appointment (sign up via Calendly link on Canvas/syllabus)
Teaching Assistant: Yiran Huang, PhD student in Earth, Environmental and Planetary Sciences
Class meetings: Tuesday & Thursday 10:30-11:50am @ Lincoln Field 120
EEPS 1340 / DATA 1340: Machine Learning for the Earth and Environment (formerly EEPS 1960D)
Course Description [flier]
This course introduces science students to modern data science tools for exploratory data analysis, predictive modeling with machine learning, and scalable algorithms for big data. Familiarize students with a cross-section of common machine learning models and algorithms emphasizing developing practical skills for working with data. Topics covered may include dimensionality reduction, clustering, time series modeling, linear regression, regularization, linear classifiers, ensemble methods, neural networks, model selection and evaluation, scalable algorithms for big data, and data ethics. The course will present case studies of these tools applied to problems in the Earth sciences. Intended audience is advanced undergraduate and graduate students in Earth, Environmental and Planetary Sciences or other physical science disciplines. Students will practice and develop their skills in data science through a hands-on project on a topic of their choice. This course is taught using the Python programming language.
Prerequisites
Required: Basic programming experience in Matlab, Python, R, or any high-level programming language (e.g. EEPS 0250, 1690, or 1430; APMA 0160; CSCI 0111, 0150, 0170, or 1090).
Recommended: Linear algebra (e.g. EEPS 1690; Math 0520, 0540; CSCI 0530) and statistics (e.g EEPS 1690; APMA 0650 or 1650; CSCI 0220) or permission of the instructor.
Students should be interested in Earth, environmental and/or planetary sciences, but prior coursework in these subjects is not required for the course.
see Resources page for review material on these topics
EEPS 1340 vs. EEPS 1720
EEPS-DATA 1340 (formerly EEPS 1960D) is an introductory data science course geared toward students in the Earth and physical sciences. EEPS 1340 assumes no prior knowledge of machine learning (ML). Students will develop hands-on data science skills through programming exercises in Python with the scikit-learn package and an applied ML project on a topic of the student's choice. The course will draw from examples in Earth, Environmental and Planetary sciences, so an interest in these topics is recommended but prior coursework in EEPS is not necessary. EEPS-DATA 1340 will be offered in Fall 2023 (anticipated next offering: Spring 2025).
EEPS-DATA 1720 (Tackling Climate Change with Machine Learning) is course designed for students with prior experience in ML (see prerequisites or consult the instructor) who are looking to deepen their knowledge of ML tools and learn about how these tools can advance climate research. Students will read and discuss recent research articles (from top climate journals and ML conferences) on ML for climate-related applications. Students will also complete a group project to gain deeper knowledge of a climate application and/or advanced ML tools. A desire to learn more about climate science and climate change solutions is required; prior coursework in climate science is not required but may be helpful. EEPS-DATA 1720 will be offered in Spring 2024 (anticipated next offering: Spring 2026).