AY 128/256: Astronomy Data Science Lab (Fall 2024)

Synopsis

This course consists of three data-centric laboratory experiments that draw on a variety of tools used by professional astronomers. Students will learn to procure and clean data (drawn from a variety of world-class astronomical facilities), assess the fidelity/quality of data, build and apply models to describe data, learn statistical and computational techniques to analyze data (e.g., Bayesian inference, machine learning, parallel computing), and effectively communicate data and associated scientific results.  This class will make use of data from facilities such as Kepler, Gaia, the Sloan Digital Sky Survey, and the Hubble Space Telescope to explore the structure and composition of the Milky Way, stars, and galaxies throughout the local and distant Universe. There is a heavy emphasis software development in the Python language, statistical techniques, and high-quality communication (e.g., written reports, oral presentations, and data visualization).

This course satisfies the Data Science major requirement for “Computational & Inferential Depth”. Read more about the upper division classes for the Data Science BA. This course also satisfies the Laboratory requirement for the Astronomy Major.

Instructors

• Joshua Bloom (Professor, joshbloom@berkeley.edu; 203 Campbell Hall)
• Aaron Parsons (Professor, aparsons@berkeley.edu; 425 Campbell Hall)
• Saahit Srihari Mogan (GSI, smogan@berkeley.edu>)
• Peter Ma (GSI, peter_ma@berkeley.edu)

Course Aims

• Introduce and motivate a range of analysis techniques and data pipelining
• Gain practical, in-depth experience doing inference on real, open-ended modern astronomical challenges
• Build reproducible, well-tested, well-documented software & infrastructure
• Learn to work with open data and code, and in an open science environment
• Hone presentation (speaking & visualization) skills
• Develop skills for future in academia, industry, …

Course Format

• 4 credits
• 2 weekly 1.5 hour meetings
• “Show & tell” progress reports + instructor lecture
• 3.5 labs
• Will require a fair amount of dedicated coding time
• Grad students will do more in-depth labs with a higher expectation for rigor

Lab Schedule:

• Lab 0 (Clusters) — Assigned: Weds 8/24, Due: Fri 9/9
• Lab 1 (Variable Stars & Dust) — Assigned: Mon 9/9, Due: Fri 10/4
• Lab 2 (Stellar Spectra) — Assigned: Mon 10/7, Due Fri 11/1
• Lab 3 (Galaxy Classification) — Assigned: Mon 11/4, Due Fri 12/6

Lectures

• Monday and Wednesday 10:30 am-noon in 131 Campbell

Office Hours

• Josh/Aaron: 12-1pm Monday (Josh) / 12-1p Friday (Aaron) (Campbell 355)
• Peter: 1-2pm Tues (355)
• Saahit: 5-7pm Wed (355) and 4-5pm Thu (233)

Grading

• 10%: Class Participation – Active engagement in class discussion and lecture, participation during “show and tell”, attending office hours, Q&A engagement on Ed Discussion (invite link)

• 65%: Lab Reports/Notebooks – due before specified class, -10% for each day late, you can collaborate with people in the class, but all work, writeups, notebooks, coding, plots, etc. MUST be your own. We’ll drop your 2 lowest checkpoint scores.

• 25%: Lab Checkpoints – weekly progress goals for each lab; graded mostly for completeness.

Prereqs

• There are no formal prereqs for this class but there are a number of informal requirements, outlined in this document, which you must satisfy.

Use of Generative AI models & LLMs

Please see the Academic Use Policy of Artificial Intelligence (AI) for the policy in this course.