DE School

LSST Dark Energy School

Four Dark Energy Schools have been held as part of LSST DESC meetings. Below you will find (most of) the materials used for the lessons presented at these schools -- slides, activities, resources -- as well as YouTube videos of the lectures.

DE School I, SLAC National Laboratory, February 2, 2015

DE School II, Argonne National Laboratory, October 26, 2015

DE School III, SLAC National Accelerator Laboratory, March 7, 2016

DE School IV, University of Oxford, July 18, 2016

DE School V, SLAC, February 13, 2017

DESC members can view the attendance lists and provide feedback at the school wiki pages:

DE School VI will be held at Stony Brook University on July 10, 2017.

DE Schools III, IV and V have been supported by the LSSTC Enabling Science effort, including matching funds from the Simonyi-Gates Challenge. All DE Schools have also received support through the host institutions.

The DE Schools promote learning through peer interactions; see A Quick Guide to Acitve Learning in Lectures for more details.


Lessons in reverse chronological order:

Cosmology with Strong Gravitational Lenses (Feb 13, 2017)

Watch the class on YouTube here!

Teacher: Phil Marshall, SLAC

Lesson Materials and Resources:

 

Machine Learning in the LSST Era (Feb 13, 2017)

Teacher: David Kirbky, UC Irvine

Watch the class on YouTube here!

Lesson Materials and Resources:

 

Weak Lensing Shear Estimation Methods and Systematics (Feb 13, 2017)

Teacher: Rachel Mandelbaum, Carnegie Mellon University

Watch the class on YouTube here!

Lesson Materials and Resources:

  • Lesson slides
  • Pre-lesson reading - Introduction (section 1) of Jarvis et al., 2016.
  • Further resources -
    • GREAT3 results paper - Various WL systematics including noise bias and the impact of realistic galaxy morphology.
    • Refregier et al 2012 - Clear explanation of noise bias.
    • Hirata et al 2004 - Supposedly about a totally different topic (intrinsic alignments), but Chris Hirata derived a whole bunch of stuff about various WL systematics using a nice clear formalism in section 3.
    • The rest of Jarvis et al 2016.

 

Null Tests: Looking for Signal in All the Wrong Places (Feb 13, 2017)

Teacher: Mike Jarvis, U of Pennsylvania

Watch the class on YouTube here!

Lesson Materials and Resources:

 

The Era of Large Surveys: What Will LSST Deliver? (July 18, 2016)

Teacher: Mario Juric, U of Washington

Watch the class on YouTube here!

Lesson Materials and Resources:

 

How Bright is that Object? (July 18, 2016)

Teacher: Robert Lupton, Princeton University

Watch the class on YouTube here!

Lesson Materials and Resources:

 

Photometric Redshifts for LSST (July 18, 2016)

Teacher: Jeff Newman, University of Pittsburgh

Watch the class on YouTube here!

Lesson Materials and Resources:

 

Cross-Correlations of Dark Energy Probes (July 18, 2016)

Teacher: Jo Dunkley, University of Oxford

Watch the class on YouTube here!

Lesson Materials and Resources:

  • Lesson slides (with notes from talk embedded)
  • Exploiting Cross Correlations and Joint Analyses, Report from the "Dark Energy and CMB" working group for the APS DPF longterm planning exercise ("Snowmass") 2013.
  • 'Looking through the lens', Schaan et al.
  • For a small part of this lesson, install the CLASS code (c and optional python wrapper) on your laptop. Find it here: http://class-code.net. As the instructions say, either download the tarball, or grab it from git. It takes < 5 minutes to install and run, just get the c code for this lesson and check it runs by doing 'make clean', 'make class', and then './class explanatory.ini'. We just need the c code this time. As the instructions say, you might need to comment out the -fopenmp flag to compile without openMP. Then also copy in this ini file: test_mpk.ini. For plotting, you need either gnuplot or other quick plotter. There is also a python plotter that comes with CLASS but we don't need this today.

 


Robust Model Fitting with Applications to Astronomy (March 7th, 2016)

Teacher: Andrew Connolly, U of Washington

Watch the class on YouTube here!

Lesson Materials and Resources:

  • Chapter 8 (Regression and Model Fitting) in  "Statistics, Data Mining, and Machine Learning in Astronomy: A Practical Python Guide for the Analysis of Survey Data" by Željko Ivezić, Andrew J. Connolly, Jacob T. VanderPlas & Alexander Gray (2014). Errata for Chapter 8 can be found here.
  • This lesson uses an iPython notebook. See DESC_school for installation instructions for python and associated packages. Example notebooks are available under the notebook tab. A tutorial for ipython is available for those new to notebooks. To download a notebook click on the title of the notebook and then on "Source" (in the top right corner). Download and test the "Histogram" notebook to ensure that you have the right packages installed. 
  • Python 2.7 is assumed for the class with the astroml, scipy, scikit-learn, and pymc packages installed. Details for installing these packages are available here.

OR

 

Primer on Wavefronts and Aberrations: the LSST Optical PSF (March 7th, 2016)

Teacher: Aaron Roodman, SLAC

Watch the class on YouTube here!

Lesson Materials and Resources:

  • Lesson slides
  • For both this lesson and the following lesson on PSFs and the atmosphere, the recommended resource is "Introduction to Fourier Optics" by Joseph W. Goodman, 3rd edition (2005), Chapters 1-4, but especially Chapter 2 (Analysis of Two-Dimensional Signals and Systems).
  • Also see this concise lecture from Observational Astronomy at U of HeidelbergDiffraction Theory (12-page pdf)
  • This lesson uses an iPython notebook. In addition to the usual packages (numpy, scipy, astropy), you need to have cython installed.  Cython is included in Anaconda; otherwise "pip install cython".
  • Download and unpack this tar file  into your working area; then issue the setup command below for the cython code:
    • tar -xvf wavetoimage.tar
    • python adaptive_moments_setup.py build_ext --inplace
  • Download this python notebook to get started: WaveToImage Dark Energy School.ipynb

 

More than Just a Phase: the LSST Atmospheric PSF (March 7th, 2016)

Teacher: Josh Meyers, Stanford University

Watch the class on YouTube here!

Lesson Materials and Resources:

 

Which Dark Energy Models Will We Test in the LSST-WFIRST-Euclid Era? (March 7th, 2016)

Teacher: Bhuv Jain, U of Pennsylvania

Watch the class on YouTube here!

Lesson Resources:

 


Dark Energy and Lensing: From GR to Data Analysis (Oct 26th, 2015)

Teacher: Scott Dodelson, Fermilab

Watch the class on YouTube here!

Overview of modified gravity models for acceleration and how they might be detected using weak lensing in LSST.
 
Learning Objectives:
  • Understand why it is difficult to understand cosmic acceleration 
  • Discuss how modified gravity models differ from dark energy models 
  • Explain gravitational lensing and how it probes the growth of structure
  • Learn how to code in cosmosis

Lesson Materials and Resources:

 

How to simulate our Universe (Oct 26th, 2015)

Teacher: Katrin Heitmann, Argonne National Laboratory

Watch the class on YouTube here! 

Lesson Materials and Resources:

 

Future Computing Architectures and Data Analysis (Oct 26th, 2015)

Teacher: Salman Habib and Adrian Pope, Argonne National Laboratory

Watch the class on YouTube here!

Lesson Materials and Resources:

 

Survey Strategy and Dark Energy Systematics (Oct 26th, 2015)

Teacher: Eric Gawiser, Rutgers

Watch the class on YouTube here!

Lesson Materials and Resources:

 


How to use statistics to describe the large scale structure of the Universe (Feb 2, 2015)

Teacher: David Kirkby, U.C. Irvine

Watch the class on YouTube here!

An introduction to large scale structure in the Universe and techniques used to measure its statistics, including the topics of co-variance, random fields and the power spectrum. 

Learning Objectives:

After completing this class, students will be able to:

  • Describe what we mean by "Large Scale Structure"
  • Sketch the relative sizes of benchmarks such as the moon and an LSST chip
  • Estimate sigma8 using a coin and a scatter plot of galaxies
  • Discuss the limitations of cosmic variance and the assumptions of homogeneity and isotropy
  • Explain how r- and k-space basis functions represent the same two-point correlations in different ways
  • Match a power spectrum to its corresponding correlation function and Gaussian random field

Lesson Materials:

 

Cosmic Co-variance (Feb 2, 2015)

Teacher: Michael Schneider, Lawrence Livermore National Laboratory

Watch the class on YouTube here!

In this lesson we will cover how we infer cosmological parameter constraints in the presence of correlated errors, including how we infer cosmological parameters from large-scale structure probes and how we combine parameter constraints from different surveys. 

Learning Objectives:

After completing this class, students will be able to:

  • Calculate the error ellipse for cosmological parameter constraints, given a covariance matrix.
  • Describe the covariance matrix for the Fourier transform and power spectrum of  a Gaussian random field.
  • Describe how to obtain the covariance matrix for the galaxy power spectrum, using Nbody simulations.
  • Evaluate different approaches to covariance estimation.

Lesson Materials:

 

How the Physics of Sensors Impacts Dark Energy Science (Feb 2, 2015)

Teacher: Chris Stubbs, Harvard

Watch the class on YouTube here!

This lesson deals with the non-idealities encountered in real-world CCDs. We illustrate how to assess the impact of one particular gremlin on weak lensing measurements, lateral electric fields arising from radially symmetrical impurities in the Silicon wafer. 

Learning Objectives:

After completing this class, students will be able to:

  • Identify the differences between ideal and real CCDs.
  • Describe how CCD features (such as tree rings, charge transport anomolies) impact LSST images, and evaluate their impact on weak lensing measurements.
  • Explain what measurements we can make in the lab to identify and correct for CCD effects.

Lesson Materials and Resources:

 

Key LSST Design Choices, and How They Were Driven by Science (Feb 2, 2015)

Teacher: Steve Kahn (LSST/SLAC)

Watch the class on YouTube here!

Learning Objectives:

After completing this class, students will be able to:

  • Identify how LSST science goals impacted design choices.
  • Explain the choice of limiting magnitude of 27.5 for the LSST survey, and the interplay between survey depth and breadth.
  • Describe how the cadence choices for LSST impact solar system science and the transient optical sky.

Lesson Materials and Resources:

 

Image processing algorithms: Building science-ready catalogs (Feb 2, 2015)

Teacher: Jim Bosch, Princeton

Watch the class on YouTube here!

A very high-level view of the LSST Data Management (DM) pipelines, and a closer look at the details of algorithms for object detection. 

Learning Objectives:

After completing this class, students will be able to:

  • Describe the LSST DM image processing pipelines
  • Explain the algorithms used for object detection in images
  • Understand how single-epoch detection extends to multiple epochs
  • Discuss detection algorithms for different science cases

Lesson Materials and Resources:

 

LSST's Cosmological Probes (Feb 2, 2015)

Teacher: Shirley Ho, Carnegie Mellon University

Watch the class on YouTube here!

This lesson is designed to introduce several of the dark energy probes in LSST. It assumes basic knowledge of cosmology and statistics. 

Learning objectives:

After this class, students will be able to:

  • Desccribe how different types of cosmological measurements are made
  • Explain how these measurements are used to infer something about our cosmological model. 

Lesson Materials and Resources: