AstroData Hack Week

University of Washington

September 15-19, 2014

Astro Hack Week is a week-long summer school / hack week / unconference focused on astrostatistics and data-intensive astronomy.

The mornings will be a typical summer school format, with lectures and exercises covering essential skills for working effectively with large astronomical datasets. The afternoons will be entirely unstructured, and offer opportunities for collaborative research, breakout sessions on special topics, and application of the concepts covered during the morning sessions.

The vision is to provide a space to encourage learning, research, collaboration, and sharing of expertise, for the benefit of both young and experienced astronomical researchers alike.

Hack Week Wiki

See our github wiki for up-to-date information on scheduling and planning, including a partial list of who will be attending and what they plan to work on!


This will be a relatively small event; we have space for about 40 people. Registration is $50 per participant, which will partially offset the costs of running the workshop.

We have reserved a block of rooms at the University Inn, a short walk from campus; the conference rate of $149 (single or double occupancy) is available first-come first-served.


The following folks have been involved with planning this event, and will be teaching the morning sessions and facilitating the afternoon hack times:

Joshua Bloom

Željko Ivezić

David W. Hogg

Daniela Huppenkothen

Phil Marshall

Fernando Perez

Jake VanderPlas


Each day will consist of morning sessions devoted to learning fundamentals of computational astronomy, and afternoon sessions devoted to unstructured research, hacking, and collaboration, with the potential for informal breakout sessions on special topics of interest.

We will be making use of the textbook Statistics, Data Mining, and Machine Learning in Astronomy by Ivezić, Connolly, VanderPlas, and Gray. It is not required, but we encourage you to bring a copy if you have it!

The following is an initial draft of the schedule; there will likely be modifications made in the weeks leading up to the conference.

Monday, September 15th

9:00am-9:30am Coffee
9:30am-12:00pm Scientific Computing with Python
  • Interactive Computing & Reproducible Research with IPython
  • Effective Computing with NumPy
  • Visualization with Matplotlib
  • Exploring computational tools available in SciPy
12:00pm-12:30pm Hack Idea Intros
12:30pm-1:30pm Lunch on-site
1:30pm-5:30pm Hack time & Breakouts
5:30pm-6:00pm Daily Wrap-up

Tuesday, September 16th

9:00am-9:30am Coffee
9:30am-12:30pm Classical Statistics & Modeling
  • Intro to classical probability theory
  • Maximum likelihood Optimization & Uncertainty Quantification
  • Goodness of Fit and Hypothesis Testing
  • Confidence Estimates using Bootstrap
12:30pm-1:30pm Lunch on-site
1:30pm-5:30pm Hack time & Breakouts
5:30pm-6:00pm Daily Wrap-up

Wednesday, September 17th

9:00am-9:30am Coffee
9:30am-12:30pm Bayesian Statistics & Modeling
  • Bayes' Theorem and Bayesian probability
  • Bayesian Priors
  • Posterior optimization, marginalization, and Uncertainty Quantification
  • Hypothesis Testing
  • Intro to Markov Chain Monte Carlo (MCMC) sampling
12:30pm-1:30pm Lunch on-site
1:30pm-5:30pm Hack time & Breakouts
5:30pm-6:00pm Daily Wrap-up

Thursday, September 18th

9:00am-9:30am Coffee
9:30am-12:30pm Supervised Machine Learning & Model Fitting
  • Supervised Machine Learning: Classification vs Regression
  • A survey of Classification techniques
  • A survey of Regression techniques
12:30pm-1:30pm Lunch on-site
1:30pm-5:30pm Hack time & Breakouts
5:30pm-6:00pm Daily Wrap-up
7:00pm-late off-site dinner and hackathon

Friday, September 19th

9:00am-9:30am Coffee
9:30am-12:30pm Unsupervised Machine Learning & Data Mining
  • Dimensionality Reduction Algorithms
  • Clustering Algorithms
  • Density Estimation Algorithms
12:30pm-1:30pm Lunch on-site
1:30pm-5:00pm Hack time & Breakouts
5:00pm-6:00pm Week Wrap-up

Because of the unique and unstructured nature of this hack week, we suggest that you come prepared! Come with a project in mind, with data to explore, with a colleague to collaborate with. This should not be a week away from work, but a week to try some new approaches to your current research topic!


This workshop is being organized as part of the Moore-Sloan Data Science Initiative, together with the University of Washington, University of California Berkeley, and New York University. It is made possible by the following sponsors: