PyHEP 2020 (virtual) Workshop

Session

Fitting & statistics

16 Jul 2020, 17:00

Conveners

Fitting & statistics

• Mariel Pettee (Yale University (US))
• Jim Pivarski (Princeton University)

Description

PACIFIC TIME ZONE SESSION 4

15h00 - 16h00 PDT, 00h00 - 01h00+1 CET, 03h30+1 - 04h30+1 IST, 06h00+1 - 07h00+1 CST, 07h00+1 - 08h00+1 JST

49. Machine learning technique for signal-background separation of nuclear interaction vertices in the CMS detector

Anna Kropivnitskaya (The University of Kansas (US))

16/07/2020, 17:00

Notebook talk

47. pyhf: Accelerating analyses and preserving likelihoods (TUTORIAL)

Matthew Feickert (Univ. Illinois at Urbana Champaign (US))

16/07/2020, 17:30

Tutorial

1. Error computation in iminuit and MINUIT: How HESSE and MINOS work

Dr Hans Peter Dembinski (Max-Planck-Institute for Nuclear Physics, Heidelberg)

Notebook talk

We review the two algorithms that compute fit uncertainties in virtually any HEP fit, HESSE and MINOS from the MINUIT project. We will discuss the theoretical foundation of these algorithms, how MINUIT implements them, and practical aspects such as computational speed. On toy examples, we will study the frequentist coverage probability of confidence intervals computed with both methods.

17. Model building and statistical inference with zfit and hepstats

Jonas Eschle (Universitaet Zuerich (CH)), Matthieu Marinangeli (EPFL - Ecole Polytechnique Federale Lausanne (CH))

Tutorial

zfit is a model fitting library based on top of TensorFlow and built for customization. It can build models, load data, create and optimize losses. hepstats is a package for statistical inference and is build on top of the zfit interface, and can therefore use models and losses built in zfit directly.

In this tutorial, we propose to split the tutorial into two parts (switching speaker...

25. SModelS -- a tool for interpreting simplified-model results from the LHC

Wolfgang Waltenberger (Austrian Academy of Sciences (AT))

Notebook talk

SModelS is an automatic, public python tool for interpreting simplified-model results from searches for new physics at the LHC. It is based on a general procedure to decompose Beyond the Standard Model (BSM) collider signatures presenting a Z2 symmetry into Simplified Model topologies. Our method provides a way to cast BSM predictions for the LHC in a model independent framework, which can be...

20. Tensorflow-based Maximum Likelihood fits for High Precision Standard Model Measurements at CMS

Josh Bendavid (CERN)

Notebook talk

With increasing integrated luminosity at the LHC, highly differential and extremely precise measurements of Standard Model processes become possible. This imposes stringent requirements for accurate modelling of systematic uncertainties of both experimental and theoretical origin. One possible method for the unfolding of detector response is the use of maximum likelihood fits, with systematic...

15. zfit - TensorFlow 2.0: dynamic and compiled HPC

Jonas Eschle (Universitaet Zuerich (CH))

Notebook talk

zfit is a model fitting library completely implemented in Python and based on the Deep Learning framework TensorFlow. With the recent release of TensorFlow 2.0, the structure of the TensorFlow library, as well as zfit, fundamentally changed; what was before a head-twisting exotic graph building library became a numpy-like, JIT compilable computational backend. It works with Numpy compatible...