Skip to content

smotherh/uw-astr598-w18

 
 

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

ASTR 598, Winter 2018, University of Washington:

Astro-statistics and Machine Learning

Andy Connolly and Željko Ivezić

Location

  • When: 11:00-12:20, Tuesday & Thursday, Winter quarter 2018
  • Where: PAB B305 (close to the end of the grad student hallway)

Class Materials

Reference textbook

Ivezić, Connolly, VanderPlas & Gray: Statistics, Data Mining, and Machine Learning in Astronomy: A Practical Python Guide for the Analysis of Survey Data (Princeton University Press, 2014)

Class Description

This course will introduce graduate students to most common statistical and computer science methods used in astronomy and other physical sciences. It will combine theoretical background with examples of data analysis based on modern astronomical datasets. Practical data analysis will be done using python tools, such as astroML module (see www.astroML.org). While focused on astronomy, this course should be useful to all graduate students interested in data analysis in physical sciences and engineering. The lectures will be aimed at graduate students and the main discussion topics will be based on Chapters 3-5, and selected topics from Chapters 6-10, from the reference textbook.

By taking this course, students will develop working knowledge of topics such as robust statistics, hypothesis testing, maximum likelihood analysis, Bayesian statistics, model parameter estimation, the goodness of fit and model selection, density estimation and clustering, unsupervised and supervised classification, dimensionality reduction, regression and time series analysis. Most of these topics will be applied in class homeworks and projects to analysis of astronomical data.

About

ASTR 598: Astro-statistics and Machine Learning

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Jupyter Notebook 94.2%
  • HTML 5.8%