*REMOTE* CMS Open Data Workshop for Theorists at the LPC

30 Sept 2020, 08:30 → 2 Oct 2020, 17:00 America/Chicago

Edgar Fernando Carrera Jarrin (Universidad San Francisco de Quito (EC)), Jesse Thaler (MIT), Kati Lassila-Perini (Helsinki Institute of Physics (FI)), Matthew Bellis (Cornell University/Siena College (US))

Feedback survey: https://forms.gle/m4Hmw8oCvSTR8Fj17

The CMS Open Data Workshop for Theorists will take place on September 30-October 2, 2020. It will be ZOOM ONLY.

In 2014, CMS released a significant amount of data through the CERN Open Data Portal, available to anyone to analyze as they saw fit. In 2017, a theory group at MIT published two peer-reviewed publications using this data, prompting renewed discussion as to how to make these open datasets easier to work with by non-CMS analysts. The goal of this workshop is to lower the threshold for access to these data for theorists and phenomenologists. Attendees will be lead through the steps of how to login to the CERN virtual machines, run the CMS analysis software, and perform the basics of an analysis. All exercises will be hands-on and participants should be prepared to dive into the data right away. Time will also be spent brainstorming with attendees about how the entire process of accessing and analyzing the data could be made more useful for the broader HEP community.

Please join us and register before the deadline of September 23rd.

Everyone has to follow Fermilab's code of conduct:

http://directorate-docdb.fnal.gov/cgi-bin/RetrieveFile?docid=174

Organizing Committee:
Matthew Bellis (Siena College)
Edgar Carrera (U. San Francisco de Quito)
Kati Lassila-Perini (U. of Helsinki)
Jesse Thaler (MIT)

Local Organizing Committee:
Gabriele Benelli (Brown U.)
Christian Herwig (Fermilab)
Julie Hogan (Bethel U. and Brown U.)
Clemens Lange (CERN)
Andrew Melo (Vanderbilt U.)
Nada Mohamed (Siena College)
Stephen Mrenna (Fermilab)
Kevin Pedro (Fermilab)
Emanuele Usai (Brown U.)
David Yu (Brown U.)

LPC Events Committee:
Gabriele Benelli (Brown U., Co-Chair)
Kevin Pedro (Fermilab, Co-Chair)

LPC Coordinators: 
Cecilia Gerber (UIC)
Sergo Jindariani (Fermilab)

GroupPicture_CMSODWS_2Oct2020.png

Wednesday, 30 September

1 Welcome and orientation

CMSOpenDataForTheoristsWorkshopLogistics_v1.pdf

jthaler_2020_09_OpenData_WorkshopWelcome.pdf

WelcomeAndOrientation_CMSODW.mp4

WelcomeAndOrientation_CMSODW_partialTranscipt.txt

2 Live Presentation: Workshop Introduction

Speaker: Kati Lassila-Perini (Helsinki Institute of Physics (FI))

CMS-OD-Workshop-2020-intro.pdf

Google Slides: Let’s introduce ourselves!

WorkshopIntroduction_CMSODWS.mp4

WorkshopIntroduction_CMSODWS_partialTranscipt.txt

09:45 Break

3 Live Hands-on lesson: Dataset scouting

This lesson is designed to teach you how to use the command-line to explore the directories where the data is stored. In this way, you can see what triggers were applied when the data was taken and what Monte Carlo samples are available for the run period you are interested in.

You’ll also be shown how to do a first-order inspection of some of these datafiles, just to see what is stored in them.

Speaker: Matthew Bellis (Cornell University/Siena College (US))

DataScouting_CMSODWS.mp4

DataScouting_CMSODWS_partialTranscipt.txt

Dataset scouting

4 Live Hands-on lesson: Trigger manipulation

In this lesson you will:

learn what the CMS trigger system is
learn how to select and understand triggers for you analysis
learn how to obtain trigger prescales and acceptance bits

Speaker: Edgar Fernando Carrera Jarrin (Universidad San Francisco de Quito (EC))

Trigger Manipulation

TriggerManipulation_CMSODWS.mp4

12:00 Break

5 Async Demo: Physics objects

In this video tutorial we will introduce you to CMS particle flow
and physics objects. Later, in the afternoon, you will dive into the code. Enjoy!

A youtube video link is attached as well as the slides presented in it.

Speaker: Julie Hogan (Brown University, Bethel University (US))

OpenDataWorkshop_Objects.pdf

Youtube link: Open Data Workshop Objects demo

13:00 Lunch

6 Async Hands-on lesson: Physics objects I

When a physicist approaches an analysis using CMS data, they typically rely on the reconstruction algorithms developed by CMS to interpret detector signals into meaningful physics objects. In code, the result of these recontruction algorithms takes the form of several C++ classes that will be introduced briefly in this lesson. The content of the C++ class reflects the nature of the physics object it represents.

In this lesson we will study several fundamental particles: muons, electrons, photons, and tau leptons. The first three particles are special in CMS, because they are reconstructed as single “particle-flow candidates”. The Particle Flow algorithm (CITE ME) combines detector signals from multiple CMS subdetector systems to categorize all energy deposits as muons, electrons, photons, neutral hadrons, or charged hadrons. Tau leptons are more complex because they are not stable and have several detector signatures that include muons, electrons, photons, and/or hadrons. In the next lesson we will approach even more complex objects such as jet and missing transverse energy.

After exploring the code elements that are common to all CMS physics objects we will look at muons, electrons, photons, and tau leptons in more detail to understand the options for identifying these particles in your analysis. The final episode (MAYBE IN SELECTION LESSON?) will show how an analyzer can combine different identification elements into selection criteria.

Speaker: Julie Hogan (Brown University, Bethel University (US))

Analyzing CMS Physics Objects

7 The Future is Open: Adventures with Public Collider Data

Fermilab employees and users can access the Zoom link below (Services login required):

https://fermipoint.fnal.gov/org/ood/LabLeadership/Shared%20Documents/Zoom%20link%20for%20colloquium.docx?d=wddecabdd5efe44ee91ba775647366a0a&csf=1&e=XzG3Ib

Please note: you will need the passcode to enter the zoom

Anyone else can obtain the Zoom link the day of the colloquium by emailing Barb Kronkow at kronkow@fnal.gov

In November 2014, the CMS experiment at the Large Hadron Collider made the unprecedented move of releasing research-grade particle physics data for unrestricted use. I am a theoretical particle physicist, and for the first time, I had access to real collision data from a cutting-edge experiment, as well as an opportunity to demonstrate the scientific value of public data access. Over the past six years, my research group has carried out a number of innovative analyses using the CMS Open Data. In this colloquium, I highlight some of our research successes as well as some of the challenges we faced using public collider data to explore physics in and beyond the Standard Model.

Speaker: Jesse Thaler (MIT)

JesseThaler_FermilabColloquium_Sep30_2020.mp4

jthaler_2020_09_OpenData_FermilabColloquium.pdf

The Future is Open: Adventures with Public Collider Data

Thursday, 1 October

8 Live Hands-on lesson: Physics objects II - Jets and MET

Jets and missing transverse energy (MET) are critical for CMS physics analyses. They are more complex than most of the objects we discussed in the previous lesson, because they are reconstructed using multiple particle-flow candidates. After all candidates have been built from the tracks and energy deposits in CMS, they can be “clustered” using a variety of algorithms into composite objects called “jets”. Missing transverse energy clusters, in a sense, all candidates in the entire detector: it is the negative vector sum of the momentum of all candidates.

In this lesson we will explore the basic utilities for jets and MET, how to identify jets that arise from interesting original particles such as bottom quarks, and possibly how to correct jets and MET for differences between data and simulation.

Speakers: Farrah Simpson (Brown University (US)), Julie Hogan (Brown University, Bethel University (US)), Nikolas Pervan (Brown University (US))

Jets and missing transverse momentum

PhysicsObjectsII_CMSODW.mp4

PhysicsObjectsII_CMSODW_partialTranscript.txt

10:00 Break

9 Live Hands-on lesson: Pre-selection and skimming

In this lesson we will review the CMS data flow and summarize the selections that have been made up through the NanoAOD production. We will also go over how to produce your own set of NanoAOD files, though this will not be required to continue with the workshop.

Speakers: Farrah Simpson (Brown University (US)), Julie Hogan (Brown University, Bethel University (US)), Nikolas Pervan (Brown University (US))

Data preselection and skimming

PreSelectionAndSkimming_CMSODWS.mp4

PreSelectionAndSkimming_CMSODWS_partialTranscript.txt

10 Live Hands-on lesson: Object ID and selection.

With physics object prepared and NanoAOD files created, we are ready to begin thinking about an actual physics analysis!

In the previous exercises, you learned how to access and store object information from an AOD file and convert the AOD file to NanoAOD. The Events tree within the NanoAOD files contains all the derived information required for many searches or measurements. We will study a search for the Higgs boson in the tau tau decay channel – you can go back to the pre-exercises to find the published paper.

Speakers: Farrah Simpson (Brown University (US)), Julie Hogan (Brown University, Bethel University (US)), Nikolas Pervan (Brown University (US))

Object ID and Event Selection

ObjectIdAndSelection_CMSODWS.mp4

11:45 Break

11 Live Hands-on lesson: Plotting and interpretation

This exercise will walk you through the process of making some basic plots, once you have skimmed some data and produced ROOT files. We’ll make these plots using ROOT and calling ROOT from python using the PyROOT module. Note, that you could choose other approaches to making plots, such as using matplotlib, but the ROOT files need to be accessed with ROOT or uproot so we’ll stick with ROOT for now.

Once we’ve made the plots, we’ll spend some time examining them and trying to interpret the distributions so that we can develop some intuition for how things look.

Speaker: Matthew Bellis (Cornell University/Siena College (US))

Plotting and interpretation

PlottingAndInterpretation_CMSODWS.mp4

PlottingAndInterpretation_CMSODWS_partialTranscript.txt

13:00 Lunch

12 Async Demo: Luminosity and data quality.

Welcome. In this lesson you will:

learn what luminosity is and why it’s important
learn how luminosity is measured
learn how to calculate luminosity

Speaker: Thomas McCauley (University of Notre Dame (US))

Luminosity

Luminosity.mp4

Youtube link: workshop lesson luminosity

13 Async Hands-on lesson: Efficiency Studies using the Tag and Probe Method

Speakers: Allan Da Silva Jales (Universidade do Estado do Rio de Janeiro (BR)), Thomas Gaehtgens

Efficiency Studies using the Tag and Probe Method

Friday, 2 October

14 Live Demo: CMS analysis in the cloud

Speakers: Adelina Eleonora Lintuluoto (University of Helsinki (FI)), Clemens Lange (CERN)

CMSAnalysisInTheCloudI_CMSODWS.mp4

CMSAnalysisInTheCloudI_CMSODWS_partialTranscript.txt

Open Data on Kubernetes.pdf

Tutorial - CMS analysis in the cloud

10:30 Break

15 Live Hands-on lesson: CMS analysis in the cloud

Speakers: Adelina Eleonora Lintuluoto (University of Helsinki (FI)), Clemens Lange (CERN)

CMSAnalysisInTheCloudII_CMSODWS.mov

CMSAnalysisInTheCloudII_CMSODWS_partialTranscript.txt

CMS Open Data using Kubernetes

11:45 Break

16 Live Hands-on lesson: Background estimation - the ABCD method

Speaker: Santeri Laurila (CERN)

Background estimation with the ABCD method

TheABCDMethod_CMSODWS.mp4

TheABCDMethod_CMSODWS_partialTranscript.txt

13:00 Lunch

17 Live Featured Demo: ADL

Speaker: Sezen Sekmen (Kyungpook National University (KR))

ADL4OpenData201002.pdf

ADL_CMSODW.mp4

ADL_CMSODW_partialTranscript.txt

ADL for HTauTau analysis

ADL web page

18 Wrap up and time for discussion