Monte Carlo (MC) simulations are extensively used for various purposes in modern high-energy physics (HEP) experiments. Precision measurements of established Standard Model processes or searches for new physics often require the collection of vast amounts of data. It is often difficult to produce MC samples containing an adequate number of events to allow for a meaningful comparison with the data, as substantial computing resources are required to produce and store such samples. One solution often employed when producing MC samples for HEP experiments is to partition the phase space of particle interactions into multiple regions and produce the MC samples separately for each region. This approach allows to adapt the size of the MC samples to the needs of physics analyses that are performed in these regions. In this paper we present a procedure for combining MC samples that overlap in phase space. The procedure is based on applying suitably chosen weights to the simulated events. We refer to the procedure as “stitching”. The paper includes different examples for applying the procedure to simulated proton-proton collisions at the CERN Large Hadron Collider.

Estonia

Karl Ehatäht

Christian Veelken

Stitching Monte Carlo samples

The analysis of vast amounts of data constitutes a major challenge in modern high energy physics experiments. Machine learning (ML) methods, typically trained on simulated data, are often employed to facilitate this task. Several choices need to be made by the user when training the ML algorithm. In addition to deciding which ML algorithm to use and choosing suitable observables as inputs, users typically need to choose among a plethora of algorithm-specific parameters. We refer to parameters that need to be chosen by the user as hyperparameters. These are to be distinguished from parameters that the ML algorithm learns autonomously during the training, without intervention by the user. The choice of hyperparameters is conventionally done manually by the user and often has a significant impact on the performance of the ML algorithm. In this paper, we explore two evolutionary algorithms: particle swarm optimization and genetic algorithm, for the purposes of performing the choice of optimal hyperparameter values in an autonomous manner. Both of these algorithms will be tested on different datasets and compared to alternative methods.

Laurits Tani

Diana Rand

Mario Kadastik

Evolutionary algorithms for hyperparameter optimization in machine learning for application in high energy physics

A dedicated search is presented for new phenomena in inclusive 8- and 10-jet final states with low missing transverse momentum, with and without identification of jets originating from b quarks. The analysis is based on data from proton–proton collisions corresponding to an integrated luminosity of 19.7fb−1 collected with the CMS detector at the LHC at s=8TeV . The dominant multijet background expectations are obtained from low jet multiplicity control samples. Data agree well with the standard model background predictions, and limits are set in several benchmark models. Colorons (axigluons) with masses between 0.6 and 0.75 (up to 1.15) TeV are excluded at 95% confidence level. Similar exclusion limits for gluinos in R -parity violating supersymmetric scenarios are from 0.6 up to 1.1 TeV. These results comprise the first experimental probe of the coloron and axigluon models in multijet final states.

Armenia

Vardan Khachatryan

Albert M Sirunyan

Armen Tumasyan

Austria

Wolfgang Adam

Ece Aşılar

Thomas Bergauer

Johannes Brandstetter

Erica Brondolin

Marko Dragicevic

Janos Erö

Martin Flechl

Markus Friedl

Rudolf Fruehwirth

Vasile Mihai Ghete

Christian Hartl

Natascha Hörmann

Josef Hrubec

Manfred Jeitler

Axel König

Ilse Krätschmer

Dietrich Liko

Takashi Matsushita

Ivan Mikulec

Dinyar Rabady

Navid Rad

Babak Rahbaran

Herbert Rohringer

Jochen Schieck

Josef Strauss

Wolfgang Treberer-Treberspurg

Wolfgang Waltenberger

Claudia-Elisabeth Wulz

Belarus

Vladimir Mossolov

Nikolai Shumeiko

Juan Suarez Gonzalez

Belgium

Sara Alderweireldt

Eddi A De Wolf

Xavier Janssen

Jasper Lauwers

Merijn Van De Klundert

Hans Van Haevermaet

Pierre Van Mechelen

Nick Van Remortel

Alex Van Spilbeeck

Shimaa Abu Zeid

Freya Blekman

Jorgen D'Hondt

Nadir Daci

Isabelle De Bruyn

Kevin Deroover

Natalie Heracleous

Steven Lowette

Seth Moortgat

Lieselotte Moreels

Annik Olbrechts

Quentin Python

Stefaan Tavernier

Walter Van Doninck

Petra Van Mulders

Isis Van Parijs

Hugues Brun

Cécile Caillol

Barbara Clerbaux

Gilles De Lentdecker

Hugo Delannoy

Giuseppe Fasanella

Laurent Favart

Reza Goldouzian

Anastasia Grebenyuk

Georgia Karapostoli

Thomas Lenzi

Alexandre Léonard

Jelena Luetic

Thierry Maerschalk

Andrey Marinov

Aidan Randle-conde

Tomislav Seva

Catherine Vander Velde

Pascal Vanlaer

Ryo Yonamine

Florian Zenoni

China

Fengwangdong Zhang

Anna Cimmino

Tom Cornelis

Didar Dobur

Alexis Fagot

Guillaume Garcia

Muhammad Gul

Deniz Poyraz

Sinem Salva Diblen

Robert Schöfbeck

Michael Tytgat

Ward Van Driessche

Efe Yazgan

Nicolas Zaganidis

Hamed Bakhshiansohi

France

Camille Beluffi

Olivier Bondu

Sébastien Brochet

Giacomo Bruno

Search for new phenomena in events with high jet multiplicity and low missing transverse momentum in proton–proton collisions at s=8TeV