Summer Students 2013

Table of contents:

Students

Jeff Picard

Ann Wang

Georges Obied

Natalie Harrison

Ian Wessen

Da An

Valère Lambert

Gautam Upadhya

Razor variable analysis of simple SUSY models

I will be studying simple SUSY models using the razor variables. In particular I will be trying to distinguish between these models and the common ttbar background process, but using a multi-b-jet analysis. Some of the models of interest include RPV-gluino-gluino, T2bw, T2tt dna T6ttHH.

Cedric Flamant

Anthony Lutz

Karla Peña

Dustin Anderson

Thomas Arnold

Machine Learning for the LHC Grid

My research will focus on exploring how machine learning, specifically reinforcement learning, can be used to better handle matching of the workflows of the LHC experiments with LHC Grid resources. I will be using the RLPypackage (written in Python) developed at MIT to facilitate the investigation into reinforcement learning by implementing my own simulations of computer systems. The aim is to understand which learning systems better fit with a distributed system as the LHC Grid as well as to evaluate the speed with which such systems can be learned.

Christopher Hayes

Marco Cruz-Heredia

Michaelangelo Lucas

Python for Reinforcement Machine Learning

Machine learning has found wide applications in systems where decisions have to be made based on data. This includes control systems, computer systems, data analysis, etc. One of the tools of choice is often the python programming language, due to the very large amount of existing python libraries as well as to the general popularity of the language itself. During my FSRI research I plan to write python modules and programs that implement machine learning scenarios in the context of simple computer systems.

Proposals

Procedure

• Please go ahead and submit your proposal for CERN summer students urgently tomorrow Feb 15.
• Link for potential supervisors to submit a project proposal(s): http://hrapps.cern.ch/auth/f?p=112:1
• Login: AIS login name and password
• Once you have completed your proposal please don't forget to hit the "create" button in order to submit your proposal as otherwise it will not be registered.

Adi Bornheim :

A high precision time-of-flight detector for CMS

(Adi, Artur, Javier, Cristian)

A time-of-flight (TOF) detector with a timing resolution of order a few pico
seconds is discussed for the CMS Phase II upgrade for the high luminosity
LHC (HL-LHC). A major challenge at HL-LHC will be the number of simultaneous
interactions per bunch crossing (pile-up, PU). An average of 140 PU interactions
is expected with the accelerator desing currently envisioned for HL-LHC.
Based on toy MC studies and fully simulated events algorithms
have to be developed which allow to associate a vertex location to a physics
object such as a Photon or a Jet. Filtering techniques need to be developed which
allow to suppress energy deposits from pile-up interactions in the object reconstruction.
The goal of the project is to work on the development of these algorithms,
test them on simulated events as well as on real CMS data.

Geant simulation for a high precision time-of-flight detector for CMS

(Cristian, Adi, Artur, Emanuele)

A time-of-flight (TOF) detector with a timing resolution of order a few pico
seconds is discussed for the CMS Phase II upgrade for the high luminosity
LHC (HL-LHC). A major challenge is to develop radiation hard detector technology
which allows the time of flight of all relevant final state particles produced in
LHC proton-proton collisions. A fundamental question is how the physical size
of a calorimetric device limits the achievable timing measurement precision.
The goal of this project is to carry out Geant as well as ray tracing simulation
based on the existing CMS ECAL crystals as well as proposed upgrade options
such as a crystal sampling calorimeter.

ECAL calibration and performance in the presence of radiation induce noise

(Adi, Emanuele)

The noise in the CMS electromagnetic calorimeter (ECAL) increases due to effects of
the radiation environment in LHC. This degrades the resolution performance and
has an impact of the isolation criterion which is used to distinguish electrons
and photons from jets. The increased noise also has an adverse effect on the ability
to select isolated low energy photons, which are used to calibrate the ECAL.
The goal of this project is to measure precisely the impact of the noise on the
ECAL performance and calibration procedure and develop a strategy to mitigate the
effect of the noise.

Emanuele Di Marco :

H->WW->2l2nu with full decay kinematics

CMS data collected at pp collisions at 7 and 8 TeV in LHC Run1
gave an observation for a Higgs-like boson with mass 126 GeV.
For this mass, due to the large BR and very clean final state, one the most
sensitive channel at CMS is H->WW->2l2nu. The problem with this decay
mode is the absence of a clear mass peak due to two undetected
neutrinos. The student should develop an alternative strategy
for the usage of available decay kinematics, in order to
separate the signal from non-resonant WW production and give
access to the mass also in this channel.
If this is achieved, the other informations of the opening
angle of the leptons can be used to determine the spin/parity
of the new particle.

H -> ZZ-> 4 leptons property measurement

CMS data collected at pp collisions at 7 and 8 TeV in LHC Run1
gave an observation for a Higgs-like boson with mass 126 GeV.
The golden mode in terms of sensitivity to the Higgs mass is H -> ZZ
-> 4 leptons final state. The student will study the applications of the full
kinematics available in this channel to characterize the properties of
a possible signal for a resonance in the still allowed mass range to
distinguish a Standard Model Higgs boson from Higgs "look-alikes".
He will also contribute to the experimental analysis using at
best the information on lepton momentum determination to improve
the determination of the mass of the new particle.

Artur Apresyan

Alex Mott

Higgs properties measurement in H to gamma gamma

CMS data collected in the LHC run 1 has established the existence of a new boson with a mass near 125 GeV, which (among other modes) decays into two photons. Because of the extremely high resolution of the CMS Electromagnetic Calorimeter (ECAL) and relatively high signal yields in this channel, it is ideal to study the properties of this newly discovered particle. The student will study the properties of this boson with the existing LHC run 1 dataset and the expected sensitivity for precision measurement with the expected LHC run 2 dataset. The student will also measure the expected improvement in sensitivity for properties measurement from various proposed upgrades to the CMS detector, which will provide an important input to CMS upgrade planning.

Yi Chen

HCAL noise

After LS1 there is a good chance we will collect data at 25ns spacing as opposed to 50ns bunch spacing with high pileup. This poses an interesting challenge for the sub-detectors to develop efficient noise filters/algorithms designed for short bunch spacing. Hadronic calorimeter is important for most physics we want to do at the LHC. As we didn't have physics runs with 25ns bunch spacing, not much has been developed so far in terms of filters. In this project we will probe different possibilities of the noise-signal rejection algorithms using collision data collected in 2012 at both 50ns bunch spacing and 25ns bunch spacing, as well as mixed samples from data to emulate condition of high pileup and short bunch spacing. It's a nice chance to do important pioneering work that will affect the way we reject noise in the next run period.

Javier Duarte

Razoy SUSY

Si Xie

Full event kinematics in H to Zgamma

CMS data collected at pp collisions at 7 and 8 TeV in LHC Run1
gave an observation for a Higgs-like boson with mass 126 GeV.
The decay channel of this boson to a Z boson and a photon has unique
sensitivity to probe for heavy new particles that are not predicted by
the Standard Model. The student will study a number of possible
improvements to the sensitivity of the search including the use of the
full event kinematics and the use of event by event mass resolution.
In addition, the student will study the possibility to use the full
event kinematics to characterize additional properties of the Higgs-like
boson.

General Information

Information from Summer 2012

Check last years twiki for further logistical info : https://twiki.hep.caltech.edu/twiki/bin/view/Main/Summer2012 Summer2012

Projects and Final Reports from previous SURF students

• Takeda, Alex and Spiropulu, M. and Xie, S. and Di Marco, E. Validation of procedures used by CMS in the characterization of Higgs candidate events
• Horton, Max and Spiropulu, M. and Duarte, J. and Rogan, C. (2011) Developing A Maximum Likelihood Fit Workspace In RooFit/ROOT
• Lee, Lisa and Spiropulu, M. and Chen, Y. and Apresyan, A. (2012) Optimization of the hadron calorimeter in the CMS detector in the search for physics beyond the Standard Model
• Lee, Lisa and Spiropulu, M. and Chen, Y. and Apresyan, A. (2012) Study of the HCAL Fit-Based Noise Filters in 2012 Data
• Chen, Kevin and Spiropulu, M. and Kcira, D. and Bunn, J. (2012) Using Map/Reduce for Analyzing Physics Data from the Large Hadron Collider
• Chun, Sung Bong and Spiropulu, M. (2012) Study of Lepton Final States in Proton-Proton Collisions with the CMS Detector at the LHC
• Lambert, Valere and Bornheim, A. and Spiropulu, M. (2012) Measurement of the CMS ECAL Performance with Z Dielectron Decay Events in 2012 Data
• Tong, Tony and Di Marco, E. and Spiropulu, M. (2012) Measurement of Z+jets/W+jets with muon final states Berends-Giele Scaling using 5 fb−1 of data at sqrt(s)=8TeV
• Takeda, Alex and Spiropulu, Maria and Di Marco, Emanuele (2012) Study of Higgs detection in CMS experiment, and of the sensitivity to a Higgs boson in the mH < 130 GeV range
• Tong, Tony and Di Marco, E. and Spiropulu, M. (2012) W+jets and Z+jets studies with 5 fb−1 of proton-proton collisions at sqrt(s) = 8 TeV