CERN School of Computing 2024

8 Sept 2024, 15:00 → 21 Sept 2024, 11:00 Europe/Zurich

Building 1b SR4a/b (DESY)

Alberto Pace (CERN), Kristina Gunne (CERN), Andrzej Nowicki (CERN), Judith Katzy (DESY, HAMBURG)

Welcome to the 45^th CERN School of Computing (CSC 2024)

The school will take place between the 8^th-21^st September 2024 in Hamburg, Germany. This year’s School is organized in collaboration with the Deutches Elektronen-Synchroton (DESY) and the event will be hosted at the DESY campus in Hamburg.

Academic Programme

The two-week programme will consist around 50 hours of lectures and hands-on exercises, covering three main themes: physics computing, software engineering, and data technologies. Students who pass the final optional exam will receive a diploma from the CSC, as well as ECTS credits.

Other activities

However, it's not all study; the social and sport programme is also a vital part of the School. We will have ample opportunities to explore and experience some of the great cultural, historical and natural attractions of Hamburg and its surroundings.

The application for this school is now closed!

Important dates

• Wednesday 14th of February - applications open
• Monday 15 April (midnight UTC+2 / CEST) - extended deadline for applications
• Monday 29 April - invitations sent to the selected participants
• Wednesday 29 May - registration fee payment deadline
• Sunday 8 September (afternoon/evening) - student arrivals at DESY, Hamburg
• Saturday 21 September (morning) - departure

Who can apply?

The School is aimed at postgraduate (ie. minimum of Bachelor degree or equivalent) students, engineers and scientists with a few years' experience in particle physics, in computing, or in related fields. We welcome applications from all countries and nationalities. Applicants are responsible for ensuring that their registration fee and travel cost is covered by their home institute or employer, or, failing this, themselves. 

 

Sunday 8 September

15:30 Arrival and registration at DESY

19:30 Dinner DESY

1 Self presentations

Getting to know fellow participants, and exchanging with them, is a major added value of our schools. To facilitate this, we will have a self-presentation session on the first day of the school. Each participant, lecturer and organiser will have a possibility to present him-/herself during one minute, using a single slide. You may want to use this opportunity to mention where you come from, your current work, your professional interests - but also your hobbies, things you've done and you're proud of (or not 🙂) etc. The slides will be visible only to other participants of the school, but not public.

Monday 9 September

2 Opening Ceremony - Welcome

3 The DESY laboratory and research

Speaker: Beate Heinemann (DESY and University of Freiburg (Germany))

4 Welcome address from the German Ministry of Science

Speaker: Martin Thome (Bundesministerium für Bildung und Forschung - BMBF)

5 Research at CERN: LHC and beyond

Speaker: Joachim Josef Mnich (CERN)

6 70 years of Physics Research and Discoveries at CERN

Speaker: Rolf Heuer (Deutsches Elektronen-Synchrotron (DE))

7 Computing Infrastructures for Research in Physics

Speaker: Enrica Maria Porcari (CERN)

8 The CERN SChool of Computing

Speaker: Alberto Pace (CERN)

10:45 Break

9 Announcements

10 Introduction to Physics Computing L1: Hadron Collider Physics

Here we will focus on the physics of particle collisions, theoretical aspects of the standard model of particle physics, its predictive power as well as its shortcomings. Experimental aspects such as collider facilities and modern particle physics experiments as well as example physics questions and corresponding data analyses will be discussed. Furthermore, the compute models with the resulting amount of recorded data and simulated Monte Carlo events will be described.

Speaker: Arnulf Quadt (Georg August Universitaet Goettingen (DE))

11 Tools and Techniques L1: Introduction

First, we discuss some of the characteristics of software projects for high energy physics, and some of the issues that arise when people want to contribute to them. We then continue with a brief introduction to software engineering from the perspective of the individual contributor, both as a formal process and how it actually affects what you do. We discuss the examples of unit testing and memory access problems.

Speaker: Prof. Bob Jacobsen

BobJTandTLecture1.pdf

13:30 Lunch

12 Tools and Techniques L2: Tools for Collaboration, Software Engineering Across the Project

We continue the track with a discussion of system performance, and what you can (and can't) to affect it. We examine tools to help with that, discussing how they work and how they can mislead. We then discuss source control as a tool for collaboration. Using examples from basic to large and advanced, we show how individual choices can affect the building of large systems.

Speaker: Prof. Bob Jacobsen

BobJTandTLecture2.pdf

13 Exercise 1: Tools and Techniques

The exercises provide some direct experience with the tools and techniques described in the Lectures. Teams of two students will work together on examples designed to show the strengths and weaknesses of various tools and approaches. Basic and advanced exercises are available so that students can work at their own level.

The very first step is find somebody to work with and sit down together. The two of you will be working together on these exercises. These have been designed to work best two people (or occasionally three), not with one!
To start the exercises, open the instructions URL (below) in your favorite web browser and follow those instructions.
Please read that index page’s instructions all the way through. We’ve put some general info at the top that you should know, and put some reference links at the bottom that you might later discover you need.

Speakers: Prof. Bob Jacobsen, Giulio Eulisse (CERN)

16:30 Coffee break

14 Exercise 2: Tools and Techniques

This is a continuation of the Tools and Techniques exercises

Speakers: Prof. Bob Jacobsen, Giulio Eulisse (CERN)

15 Transport to dinner venue (bus)

19:30 Welcome dinner

16 Transport back to DESY

Tuesday 10 September

17 Introduction to Physics Computing L2: Digital Data, Simulation and Reconstruction in Modern Particle Physics Experiments

Here, a focus will be placed on specific detector sub-components and their data readout concepts as well as data reconstruction techniques, simulation techniques and analysis techniques.

Speaker: Arnulf Quadt (Georg August Universitaet Goettingen (DE))

18 Software Security L1: Introduction

The first lecture starts with a definition of computer security and an explanation of why it is so difficult to achieve. The lecture highlights the importance of proper threat modelling and risk assessment. It then presents three complementary methods of mitigating threats: protection, detection, reaction; and tries to prove that security through obscurity is not a good choice.

Speaker: Sebastian Lopienski (CERN)

19 Announcements

11:00 Coffee break

20 Software Design L1: Parallelism in a Modern HEP Data Processing Framework

Even though the miniaturization of transistors on chips continues like predicted by Moore's law, computer hardware starts to face scaling issues, so-called performance 'walls'. Probably, the best known is the 'power wall', which limits clock frequencies. Amongst others, a way of increasing processor performance remains now to integrate many cores in the same chip. At the same time, the upcoming LHC upgrade will increase the required CPU power drastically. Both problems challenge the current way of software design in high energy physics (HEP). Developers in high energy physics are forced to re-think their ways of software design and need to move to massively parallel applications. This lecture will explain the current HEP software design, the hardware and physics issues that need to be tackled, and possible approaches to achieve the required level of parallelization.

Speakers: Andrei Gheata (CERN), Stephan Hageboeck (CERN)

21 Software Security L2: Security in different phases of software development

The second lecture addresses the following question: how to create secure software? It introduces the main security principles (like least privilege, or defense in depth) and discusses security in different phases of the software development cycle. The emphasis is put on the implementation part: most common pitfalls and security bugs are listed, followed by advice on best practice for security development.

Speaker: Sebastian Lopienski (CERN)

13:30 Lunch

22 Study or sports time

16:00 Coffee break

23 Speeding up MadGraph5 with GPUs

The simulation of particle collisions at the Large Hadron Collider (LHC) is one of the most computing-intensive tasks in high-energy physics (HEP). The advent of High-Luminosity LHC is going to increase the demand for computing resources even more. In my lightning talk, I will discuss how we employ GPUs to speed up the MadGraph5_aMC@NLO event generator, one of the most widely used tools in HEP to simulate the initial collisions of particles, showing the main results after profiling the code and the speedups achieved by the optimization.

Speaker: Daniele Massaro (CERN)

24 The search of magnetic monopoles in the CMS Experiment

The magnetic monopoles in the CMS Experiment could be studied as a long-lived particle by the reconstruction of the trajectory of a highly ionising particle in the Tracking system with a dedicated algorithm called TrackCombiner, and the expected specific shower shape in the ECAL crystals. The processes Drell-Yan and Photon Fusion are benchmarks models for the production of magnetic monopoles in particle colliders. Both models were used to produce mass points between 1000 GeV and 4500 GeV. A 95% confidence level upper limit on the cross section of the magnetic monopole’s production via Drell-Yan process can be set for the Run 2 of the LHC.

Speaker: Thales Menezes De Oliveira (CBPF - Brazilian Center for Physics Research (BR))

25 Exercises 1: Software Security

In the practice session, a range of typical security vulnerabilities will be presented. The goal is to learn how they can be exploited (for privilege escalation, data confidentiality compromise etc.), how to correct them, and how to avoid them in the first place! Students will be given small pieces of source code in different programming languages, and will be asked to find vulnerabilities and fix them. The online course documentation will gradually reveal more and more information to help students in this task. Additionally, students will have a chance to try several source code analysis tools, and see how such tools can help them find functionality bugs and security vulnerabilities.

Speaker: Sebastian Lopienski (CERN)

26 Exercises 2: Software Security

In the practice session, a range of typical security vulnerabilities will be presented. The goal is to learn how they can be exploited (for privilege escalation, data confidentiality compromise etc.), how to correct them, and how to avoid them in the first place! Students will be given small pieces of source code in different programming languages, and will be asked to find vulnerabilities and fix them. The online course documentation will gradually reveal more and more information to help students in this task. Additionally, students will have a chance to try several source code analysis tools, and see how such tools can help them find functionality bugs and security vulnerabilities.

Speaker: Sebastian Lopienski (CERN)

19:30 Dinner at DESY

27 Pub quiz at DESY

Wednesday 11 September

28 Data Science L1: Tools for interactive data exploration

High energy physics has a rich history of interactive exploration of physics data, starting with tools like PAW and ROOT. The explosion of Data Science has created new tools for interactive exploration of large and ad-hoc datasets. This lecture introduces some of these, and shows how they can be used to find new and useful features starting with available data.

Speakers: Prof. Bob Jacobsen, Giulio Eulisse (CERN)

29 Software Design L2: Base Concepts of Parallel Programming: A Pragmatic Approach

This and the following lecture will explain the concepts behind various parallelization methodologies. First, a theoretical introduction to threads, thread-safety and concurrent data access will be given. As the recent C++ standards (starting from C++11) provide build-in support for parallel programming, their most commonly used concurrency features will be shown. Finally, concrete solutions for problems specific to concurrent programming will be discussed.

Speakers: Andrei Gheata (CERN), Stephan Hageboeck (CERN)

30 Announcements

11:00 Coffee break

31 Software Design L3: Understanding, Debugging and Profiling a Complex Multithreaded Application

Dealing with a parallel application is complex. We need to use procedures to rise fences to protect against mistakes, like static analysis tools allowing to find bugs in an automatic way. We also need to use tools to inspect and manipulate the behavior of programs at runtime, like the GDB debugger. Finally, profilers such as igprof can help us understand the performance bottlenecks of an application and get more insight on its efficiency. The objective of this lecture is to become familiar with these tools and be able to apply them in multithreaded programs.

Speakers: Andrei Gheata (CERN), Stephan Hageboeck (CERN)

32 Data Science L2: Interactive exploration of non-numeric data

This lecture continues the exploration of interactive tools, using a learn-by-doing approach. It covers approaches for statistical simulation, geographical analysis, and textual data.

Speakers: Prof. Bob Jacobsen, Giulio Eulisse (CERN)

BobJ_Lecture_DS.pdf

13:30 Lunch

33 Study or sports time

16:00 Coffee break

34 Downstream and T-Track reconstruction at the first level of LHCb trigger

SM and many new physics scenarios predict existing of Long-living particles (LLPs). However, their reconstruction is very challenging at LHC due to significantly displaced vertices. Because of the recently renovated LHCb detector and usage of GPU-based first level of the trigger, it became possible to develop two algorithms for fast track reconstruction without the usage of the hits from the first tracker (VErtex LOcator): downstream (for tracks with hits in two trackers, UT and SciFi) and faraway (for tracks with hits only in SciFi) reconstruction. Both algorithms actively use Neural networks (NN) for selection purposes. Besides serving to calibrate and align the detectors with Ks and L0 particles, the Downstream algorithm will largely increase the LHCb physics potential during Run3.

Speaker: Volodymyr Svintozelskyi (Univ. of Valencia and CSIC (ES))

35 Error underestimation in high-statistics counting experiments with finite Monte Carlo samples

Often in HEP analyses a distribution of high-statistics data is used to constrain the parameters of a complex model, containing the Physics parameter of interest (POI), but also several nuisance parameters describing systematic uncertainties. The model is estimated from Monte Carlo (MC) simulated samples and we show that their finite size can lead to the underestimation of the error on the POI in a way that is not accounted for by the traditional methods to correct for finite MC statistics. Additionally, we propose a way to estimate the size of this underestimation.

Speaker: Cristina-Andreea Alexe (Scuola Normale Superiore & INFN Pisa (IT))

36 Exercise 3: Tools and Techniques

The exercises provide some direct experience with the tools and techniques described in the Lectures. Teams of two students will work together on examples designed to show the strengths and weaknesses of various tools and approaches. Basic and advanced exercises are available so that students can work at their own level.

Speakers: Prof. Bob Jacobsen, Giulio Eulisse (CERN)

37 Exercise 3: Software Security

In the practice session, a range of typical security vulnerabilities will be presented. The goal is to learn how they can be exploited (for privilege escalation, data confidentiality compromise etc.), how to correct them, and how to avoid them in the first place! Students will be given small pieces of source code in different programming languages, and will be asked to find vulnerabilities and fix them. The online course documentation will gradually reveal more and more information to help students in this task. Additionally, students will have a chance to try several source code analysis tools, and see how such tools can help them find functionality bugs and security vulnerabilities.

Speaker: Sebastian Lopienski (CERN)

19:30 Dinner DESY

Thursday 12 September

38 Data Management L1: Setting the scene: Storage technologies, Storage reliability

The lecture presents the various Storage Models, and the supporting management techniques.

The lecture will then go in details on techniques to deliver arbitrary reliability and performance and discuss the solutions for long data preservation and trading between reliability, performances and costs.

Speaker: Alberto Pace (CERN)

39 Software Security L3: Web application security, exercise debriefing

This third hour consists of a debriefing of the exercises, and in particular those web-related. Various vulnerabilities typical to web applications (such as Cross-site scripting, SQL injection, cross-site request forgery etc.) are introduced and discussed.

Speaker: Sebastian Lopienski (CERN)

40 Announcements

11:00 Coffee break

41 Data Management L2: Cryptography, authentication, authorization and accounting 1

This lectures give elements of computer security that are relevant to data management. The lectures address the various cryptographic technologies used in data storage systems to ensure data encryption, integrity, confidentiality and access control. The Public Key infrastructure standard will be described as an example.

Speaker: Alberto Pace (CERN)

42 Exercise 4:

Speakers: Prof. Bob Jacobsen, Giulio Eulisse (CERN)

13:30 Lunch

43 Photo

44 Transport to Hamburg

45 Hamburg hafenrundfart visit excursion

46 Transport to restaurant by train

19:30 Pizza Dinner

47 Return to DESY or evening out...

Friday 13 September

48 Software Design L4: Patterns for Parallel Software Development

This lecture will present a set of common patterns in parallel programming. The sequential origin of these patterns will be discussed, as well as the restrictions that they impose. A particularly successful combination of patterns, Map-Reduce, will be described in detail and examples of its everyday use at large scale will be given. On the other hand, it will be shown how high-level features like C++ lambdas, the TBB library or the Spark framework can help get started with the aforementioned parallel patterns.

Speakers: Andrei Gheata (CERN), Stephan Hageboeck (CERN)

49 Data Management L3: Cryptography, authentication, authorization and accounting 2

This lecture will continue the discussion on various authentication technologies and then move to authorization. Accounting will also be addressed.

Speaker: Alberto Pace (CERN)

50 Announcements

11:00 Coffee break

51 Exercises 1: Software Design

The exercises will cover the topics of lectures 1 to 4 at a hands-on basis, based on C++11, TBB and Spark. It covers examples for the new C++11 functionality related to threads and thread safety. In addition, there will be examples for concurrent access to data, lock and lock-free data structures, and task based programming.  Finally, there will be an exercise to practise the Map-Reduce pattern by using the Spark parallel data processing framework.

Speakers: Andrei Gheata (CERN), Stephan Hageboeck (CERN)

52 Exercises 2: Software Design

The exercises will cover the topics of lectures 1 to 4 at a hands-on basis, based on C++11, TBB and Spark. It covers examples for the new C++11 functionality related to threads and thread safety. In addition, there will be examples for concurrent access to data, lock and lock-free data structures, and task based programming.  Finally, there will be an exercise to practise the Map-Reduce pattern by using the Spark parallel data processing framework.

Speakers: Andrei Gheata (CERN), Stephan Hageboeck (CERN)

13:30 Lunch

53 Study or sports time

16:00 Coffee break

54 Developing Artificial Intelligence in the Cloud: the AI_INFN platform & InterLink

The INFN CSN5-funded project AI_INFN ("Artificial Intelligence at INFN") aims to promote ML and AI adoption within INFN by providing comprehensive support, including state-of-the-art hardware and cloud-native solutions within INFN Cloud. This facilitates efficient sharing of hardware accelerators without hindering the institute's diverse research activities. AI_INFN advances from a Virtual-Machine-based model to a flexible Kubernetes-based platform, offering features such as JWT-based authentication, JupyterHub multitenant interface, distributed filesystem, customizable conda environments, and specialized monitoring and accounting systems. It also enables offloading mechanisms using Virtual Kubelet and InterLink API, synergizing with InterTwin. This setup can manage workflows across various providers and hardware types, which is crucial for scientific use cases that require dedicated infrastructures for different parts of the workload. The project aims to perform functional tests and benchmarks to validate its production applicability. Initial test results, emerging case studies and integration scenarios will be presented.

Speaker: Rosa Petrini (INFN Sezione di Pisa, Universita' e Scuola Normale Superiore, P)

55 Primer to Cloud Security

This talk will provide an overview of cloud security, highlighting key concepts, best practices, and methods for securing cloud-based environments. I will also share our experiences and practices for securing Kubernetes clusters at CERN. By the end of the session, attendees will gain a better understanding of fundamental cloud security principles and practical insights into managing cloud infrastructure effectively.

Speaker: Ankur Kothiwal

56 Exercise 3: Software Design

This is a continuation of the Software Design exercises.

Speakers: Andrei Gheata (CERN), Stephan Hageboeck (CERN)

57 Exercise 4: Software Design

This is a continuation of the Software Design exercises.

Speakers: Andrei Gheata (CERN), Stephan Hageboeck (CERN)

19:30 Dinner DESY

Saturday 14 September

58 Airbus Visit or free time

12:45 Picknick lunch

59 Free time

60 Free time

19:30 Dinner DESY

Sunday 15 September

61 Announcements

62 Data Analysis L1: Introduction

In this lecture we will explain what are the main goals of data analysis. We will introduce statistics as the powerful mathematical tool for data analysis. We will define probability and random variables as key concepts in statistics for data analysis.

Speaker: Toni Sculac (University of Split Faculty of Science (HR))

63 Data Analysis L2: Probability density functions and Monte Carlo methods

In this lecture we will discuss what probability density functions (PDFs) are, and what are their main properties. We will mention the most important PDFs and their properties both for discrete and continuous random variables. The importance of the Gaussian distribution lies in the Central Limit Theorem that will be discussed. Finally, we will discuss the concept of Monte Carlo methods and their usage in High Energy Physics and Data Analysis.

Speaker: Toni Sculac (University of Split Faculty of Science (HR))

64 Sunday Excursion (incl lunch)

18:30 Hamburger party

Monday 16 September

65 Data Analysis L3: Parameter estimation

In this lecture we will introduce the concept of test statistics and estimators. We will explain what are the key properties of a good estimator and how to obtain it using the Maximum Likelihood and Least Squares methods.

Speaker: Toni Sculac (University of Split Faculty of Science (HR))

66 Introduction to Machine Learning 1

Speaker: Judith Katzy (DESY, HAMBURG)

67 Announcements

11:00 Coffee break

68 Data Analysis L4: Confidence intervals

In this lecture we will define confidence intervals and make a strong statement on their statistical interpretation when discussing scientific results. We will discuss Neiman confidence belt and Bayesian confidence intervals. Finally, we will learn how to derive confidence intervals for the Maximum likelihood and Least Squares methods.

Speaker: Toni Sculac (University of Split Faculty of Science (HR))

69 Sustainable computing

Speaker: Ana Lucia Varbanescu (University of Twente)

13:30 Lunch

70 Study or sports time

16:00 Coffee break

71 The Detector Safety System, a.k.a. the reason why the detectors in the biggest accelerator in the world don't set on fire.

As the main developer of the supervision layer of the Detector Safety System (DSS) I will present what DSS is, how it works, what it looks like and some challenges in maintaining and deploying such a critical system.

Speaker: Andrea Germinario

72 Machine Learning Methods in high jet multiplicities

The four-top quarks (4tops) production in proton-proton collisions is one of the highest energy processes accessible at LHC and it was first observed by ATLAS and CMS collaborations in 2023. At ATLAS, it was measured in decay channels containing at least one leptonic decay of top quark, while leaving the all-hadronic channel unexplored. The 4tops production is very sensitive to Beyond Standard Model phenomena, therefore it is important to explore it also in the all-hadronic channel. This channel is characterised by very high jet count - 12. In this talk the machine-learning approaches for Signal and Background separation, kinematic reconstruction of top quarks, data-driven background estimation studies will be presented. SPANet, a Symmetry Preserving Attention Networks is explored. This is a promising tool for fast offline reconstruction of event topologies.

Speaker: Denys Timoshyn (Charles University (CZ))

73 HHFramework - A common framework for HH analyses in the ATLAS experiment

In preparation for the Run 3 HH analyses in the ATLAS experiment, the DiHiggs group aims to harmonize data analysis techniques, object selection, and tools into a general HH framework. This framework is designed to simplify and streamline common steps in various analyses (e.g., ntuple production, statistical analysis) while addressing the specific needs of each individual analysis (e.g., object selection, background estimation). This harmonization will also be crucial for future combinations of the various channels. In the HH->bbγγ analysis, we contribute to and utilize the integration of advanced tools within the common framework to explore the Higgs boson self-coupling and, in doing so, extend our understanding of its potential.

Speaker: Spyridon Merianos (Aristotle University of Thessaloniki (GR))

74 Exercises 1: Data Analysis

There will be 4 sets of exercises covering basic properties of PDFs and Monte Carlo generators, Maximum Likelihood fit, and (Advanced) Hypothesis testing. Students will be given realistic but simplified problems where key concepts from statistics need to be applied in order to provide scientific interpretation of data. Each set of exercises consists of 5 problems that will help guide the student. There are clear instructions to help you choose problems based on your pre-knowledge level. Data is provided in a simple text file and can be analysed with any programming language that offers libraries for statistical analysis (Python or C++ are recommended).

Speaker: Toni Sculac (University of Split Faculty of Science (HR))

75 Exercises 2: Data Analysis

There will be 4 sets of exercises covering basic properties of PDFs and Monte Carlo generators, Maximum Likelihood fit, and (Advanced) Hypothesis testing. Students will be given realistic but simplified problems where key concepts from statistics need to be applied in order to provide scientific interpretation of data. Each set of exercises consists of 5 problems that will help guide the student. There are clear instructions to help you choose problems based on your pre-knowledge level. Data is provided in a simple text file and can be analysed with any programming language that offers libraries for statistical analysis (Python or C++ are recommended).

Speaker: Toni Sculac (University of Split Faculty of Science (HR))

19:30 Dinner DESY

Tuesday 17 September

76 Introduction to machine learning 2

Speaker: Judith Katzy (DESY, HAMBURG)

77 Introduction to Machine Learning 3

Speaker: Judith Katzy (DESY, HAMBURG)

78 Announcements

11:00 Coffee break

79 Sustainable computing 2

Speaker: Ana Lucia Varbanescu (University of Twente)

80 Data Management L4: Distributed Hash Tables, Data Replication, Caching, Monitoring, Alarms and Quota 1

This lecture describes the various possible technologies used to implement distributed hash tables, data workflows and complex data transfer processes. It also discusses problems with data caching and Garbage Collection to conclude on monitoring and quota enforcement.

Speaker: Alberto Pace (CERN)

13:30 Lunch

81 Study or sports time

16:00 Coffee break

82 Negative Weights in Monte Carlo Samples.

Monte Carlo methods are indispensable tools in computational physics. However, one of the challenges encountered in Monte Carlo simulations is the occurrence of negative weights in samples. In this talk, we delve into the origins of negative weights in Monte Carlo samples, exploring their implications and practical significance. We discuss strategies to mitigate the effects of negative weights. Specifically, we focus on two approaches: optimizing event generation processes and employing cell-resampling techniques. The latter approach, which I actively utilize in my research, involves redistributing samples within discrete regions (cells) to maintain statistical integrity and mitigate the impact of negative weights.

Speaker: Nikita Dolganov (Simon Fraser University (CA))

83 FLUKA simulation for a 10 TeV Muon Collider

Circular muon colliders offer unmatched opportunities for achieving exceptionally high centre-of-mass energies. However, the continuous decay of stored muons along their trajectory presents significant technological issues for the collider and detector design. In this presentation I will discuss what are the main radiation challenges for different part of the accelerator complex.

Speaker: Daniele Calzolari (Universita e INFN, Padova (IT))

84 Exercice 1: Data Technologies

The first part of hands-on exercises aims to improve understanding of basic parameters in IO systems:
• network and media latency
• access patterns
• OS caching
• bottlenecks and optimization strategies for local and remote data access.
Few essential Linux tools will be introduced to monitor and measure IO performance avoiding bias introduced by OS caching. Students will experience and measure the impact of latency and access patterns on IO performance.
The second part covers the concept of parallelism and redundancy in storage system. We will apply the technology of Cloud storage systems to store and retrieve files in our local desktop cluster using a distributed hash table to locate files or file fragments and a REST interface to do GET, PUT or DELETE operations on these.
The exercises conclude with the implementation and performance tuning of a RAID verification algorithm.

Speaker: Andreas Joachim Peters (CERN)

85 Exercises 2: Data Technologies

The first part of hands-on exercises aims to improve understanding of basic parameters in IO systems:
• network and media latency
• access patterns
• OS caching
• bottlenecks and optimization strategies for local and remote data access.
Few essential Linux tools will be introduced to monitor and measure IO performance avoiding bias introduced by OS caching. Students will experience and measure the impact of latency and access patterns on IO performance.
The second part covers the concept of parallelism and redundancy in storage system. We will apply the technology of Cloud storage systems to store and retrieve files in our local desktop cluster using a distributed hash table to locate files or file fragments and a REST interface to do GET, PUT or DELETE operations on these.
The exercises conclude with the implementation and performance tuning of a RAID verification algorithm.

Speaker: Andreas Joachim Peters (CERN)

19:30 Dinner DESY

Wednesday 18 September

86 Introduction to Machine Learning 4

Speaker: Judith Katzy (DESY, HAMBURG)

87 Data Analysis L5 - Hypothesis testing and p-value

We will learn about the hypothesis testing procedure and all of its key concepts. We will discuss how to choose a critical region and learn about errors of first and second kind. We will learn the importance of a blinded analysis and understand all the needed steps before looking at the data. Finally, we will discuss when can we announce a discovery in science and the concept of a p-value.

Speaker: Toni Sculac (University of Split Faculty of Science (HR))

88 Announcements

11:00 Coffee break

89 Exercises 3: Data Technologies

This is a continuation of the Data Technologies exercises.

Speaker: Andreas Joachim Peters (CERN)

90 Exercises 4: Data Technologies

This is a continuation of the Data Technologies exercises.

Speaker: Andreas Joachim Peters (CERN)

13:30 Lunch

91 Study or sports time

16:00 Coffee break

92 Advanced Text Analysis for Knowledge Discovery

Advances in the generation, processing, and understanding of natural language have opened up new possibilities for extracting information from text to generate hypotheses and support scientific discovery. Because of the constantly growing amount of scientific literature, there is a real need to make it easier for experts like medical doctors to access resources published around the world and quickly draw conclusions from them. This objective can be achieved with intelligent tools for the mass searching of scientific literature texts. The research aims to explore these methods and test them on a specific corpus of literature.

Speaker: Mrs Aleksandra Kowalczuk (Accenture/University of Warsaw/WSB Merito)

93 Deep learning metrics for protein-protein interfaces in macromolecular assemblies

Structural models of macromolecular assemblies allow us to determine their function, yet the structural models we rely on are not direct experimental results, but the computational interpretation of many different noisy observations. This risks local errors that can propagate throughout the model. One area where this error can propagate is the interface between different protein chains. Can we use ML as a way to validate protein-protein interfaces and build stronger structures?

Speaker: Nicholas Whyatt (STFC UKRI)

94 Machine Learning exercise 1

Speaker: Peter Steinbach

95 Machine Learning exercises 2

Speaker: Peter Steinbach

96 CERN, Computing and talent aquisition

Software and Computing at CERN for the LHC and Beyond
CERN Opportunities for Students and Postgraduates
CERN Q&A Session
Event website

Speakers: Barbara Binder (CERN), Heather Gray (UC Berkeley/LBNL)

20:30 Networking reception

Thursday 19 September

97 Introduction to Machine Learning 5

Speaker: Judith Katzy (DESY, HAMBURG)

98 Machine Learning exercise 3

Speaker: Peter Steinbach

99 Announcements

11:00 Coffee break

100 Exercise 3: Data Analysis

There will be 4 sets of exercises covering basic properties of PDFs and Monte Carlo generators, Maximum Likelihood fit, and (Advanced) Hypothesis testing. Students will be given realistic but simplified problems where key concepts from statistics need to be applied in order to provide scientific interpretation of data. Each set of exercises consists of 5 problems that will help guide the student. There are clear instructions to help you choose problems based on your pre-knowledge level. Data is provided in a simple text file and can be analysed with any programming language that offers libraries for statistical analysis (Python or C++ are recommended).

Speaker: Toni Sculac (University of Split Faculty of Science (HR))

101 Exercise 4: Data Analysis

There will be 4 sets of exercises covering basic properties of PDFs and Monte Carlo generators, Maximum Likelihood fit, and (Advanced) Hypothesis testing. Students will be given realistic but simplified problems where key concepts from statistics need to be applied in order to provide scientific interpretation of data. Each set of exercises consists of 5 problems that will help guide the student. There are clear instructions to help you choose problems based on your pre-knowledge level. Data is provided in a simple text file and can be analysed with any programming language that offers libraries for statistical analysis (Python or C++ are recommended).

Speaker: Toni Sculac (University of Split Faculty of Science (HR))

13:30 Lunch

102 Study or sports time

16:00 Coffee break

103 Machine Learning exercise 4

Speaker: Peter Steinbach

104 Spare / DESY visit

19:30 Dinner DESY

Friday 20 September

105 Exam

11:00 Coffee break

106 Traditional football match

13:30 Lunch

107 Graduation ceremony

16:00 Free time

108 Transport to dinner venue

19:30 Closing Dinner Party

Saturday 21 September

109 Breakfast and check out