Scribd Logo
Upload

Welcome to Scribd!

  • 47 views

    Data Science and Machine Learning Workshop: M. Gonzalez-Berges M. Lonza

    Uploaded by

    Abokhaled AL-ashmawi
    The document discusses the increasing interest and applications of machine learning and artificial intelligence. It provides context on the history and evolution of machine learning, from early work in the field to recent growth enabled by increased data and computing power. Examples of machine learning techniques are presented. The document also outlines how scientific communities like HEP and accelerators are beginning to explore machine learning applications for challenges like anomaly detection, diagnostics, optimization and control of complex systems that generate large amounts of data.

    Copyright:

    © All Rights Reserved
    Download as PPT, PDF, TXT or read online from Scribd

    Data Science and Machine Learning Workshop: M. Gonzalez-Berges M. Lonza

    Uploaded by

    Abokhaled AL-ashmawi
    47 views17 pages
    The document discusses the increasing interest and applications of machine learning and artificial intelligence. It provides context on the history and evolution of machine learning, from early work in the field to recent growth enabled by increased data and computing power. Examples of machine learning techniques are presented. The document also outlines how scientific communities like HEP and accelerators are beginning to explore machine learning applications for challenges like anomaly detection, diagnostics, optimization and control of complex systems that generate large amounts of data.

    Original Title

    ML_WorkshopIntro_v3.0

    Copyright

    © © All Rights Reserved

    Available Formats

    PPT, PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    The document discusses the increasing interest and applications of machine learning and artificial intelligence. It provides context on the history and evolution of machine learning, from early work in the field to recent growth enabled by increased data and computing power. Examples of machine learning techniques are presented. The document also outlines how scientific communities like HEP and accelerators are beginning to explore machine learning applications for challenges like anomaly detection, diagnostics, optimization and control of complex systems that generate large amounts of data.

    Copyright:

    © All Rights Reserved
    Download as PPT, PDF, TXT or read online from Scribd
    Download as ppt, pdf, or txt
    47 views17 pages

    Data Science and Machine Learning Workshop: M. Gonzalez-Berges M. Lonza

    Uploaded by

    Abokhaled AL-ashmawi
    The document discusses the increasing interest and applications of machine learning and artificial intelligence. It provides context on the history and evolution of machine learning, from early work in the field to recent growth enabled by increased data and computing power. Examples of machine learning techniques are presented. The document also outlines how scientific communities like HEP and accelerators are beginning to explore machine learning applications for challenges like anomaly detection, diagnostics, optimization and control of complex systems that generate large amounts of data.

    Copyright:

    © All Rights Reserved
    Download as PPT, PDF, TXT or read online from Scribd
    Download as ppt, pdf, or txt
    You are on page 1of 17

    Introduction

    Data Science and Machine Learning


    Workshop
    M. Gonzalez-Berges
    M. Lonza

    1
    Data Science and Machine Learning Workshop - October 6, 2019 ICALEPCS2019 - New York, October 5-11, 2019
    The Control Room '70s
    the Analog Age

    2
    Data Science and Machine Learning Workshop - October 6, 2019 ICALEPCS2019 - New York, October 5-11, 2019
    The Control Room '80s
    the Digital Age

    Data Science and Machine Learning Workshop - October 6, 2019 ICALEPCS2019 - New York, October 5-11, 2019
    The Control Room '90- now
    the Communication Age

    Data Science and Machine Learning Workshop - October 6, 2019 ICALEPCS2019 - New York, October 5-11, 2019
    Control Room in the future
    the Cognitive Age

    Human-Machine Interaction
    Automation
    Artificial (Augmented) Intelligence
    5
    Data Science and Machine Learning Workshop - October 6, 2019 ICALEPCS2019 - New York, October 5-11, 2019
    Increasing interest in AI

    •European Commission view of Artificial Intelligence (Joint Research Centre - JRC):


    "The digital transformation of society has just begun: AI is central to this change and offers major
    opportunities to improve our lives"

    •USA is the first country to have developed a strategic plan for AI in 2016
    •China has started a Development Plan for AI with the goal to become the world leader in AI by 2030

    •USA Government Investments (Science Foundation, DARPA and Department for Transportation): US$
    5.3 billion
    •MIT (Massachusetts Institute of Technology) has announced an investment of US$ 1 billion and the
    creation of 50 new chairs
    •European Union has decided to increase the budget for AI in Horizon 2020 up to € 1.5 billion
    •UK announced a new investment of € 1 billion, Germany € 3 billion, France € 1.5 billion, ...

    but !!
    •Amazon investments in R&D: US$ 16.1billion in 2017

    •Alibaba, will invest in AI US$ 15 billion in three years

    6
    Data Science and Machine Learning Workshop - October 6, 2019 ICALEPCS2019 - New York, October 5-11, 2019
    Some definitions
    "Data Science is a multi-disciplinary field that uses scientific methods,
    processes, algorithms and systems to extract knowledge and insights from
    data"

    "Machine Learning is an application of Artificial Intelligence (AI) that


    provides systems the ability to automatically learn and improve from
    experience without being explicitly programmed"

    Data Science and Machine Learning Workshop - October 6, 2019 ICALEPCS2019 - New York, October 5-11, 2019
    Our Expectations on Machine Learning

    What do we want Data Science and Machine Learning to do?

    help engineers, physicist and operators to understand and manage complicated


    machines and physics phenomena often generating big amounts of data

    Examples:
    • anomaly detection
    • virtual diagnostics
    • machine tuning
    • performance optimization
    • machine models
    • ....

    8
    Data Science and Machine Learning Workshop - October 6, 2019 ICALEPCS2019 - New York, October 5-11, 2019
    Techniques used in ML applications

    https://www.linkedin.com/pulse/business-intelligence-its-relation
    ship-big-data-geekstyle/
    9
    Data Science and Machine Learning Workshop - October 6, 2019 ICALEPCS2019 - New York, October 5-11, 2019
    A quick history of Machine Learning

    10
    Data Science and Machine Learning Workshop - October 6, 2019 ICALEPCS2019 - New York, October 5-11, 2019
    The "AI Winters"

    11
    Data Science and Machine Learning Workshop - October 6, 2019 ICALEPCS2019 - New York, October 5-11, 2019
    Why ML now?

    …and many more


    12
    Data Science and Machine Learning Workshop - October 6, 2019 ICALEPCS2019 - New York, October 5-11, 2019
    Scientific Communities looking at ML

     HEP Experiments
    • Established since some years
    • Several results obtained
    • Presentations at ML conferences
    • E.g. https://iml.web.cern.ch/

     Accelerators
    • Started recently
    • Some workshops
    − Intelligent Controls for Particle Accelerators (Daresbury Lab, 2018)
    − 2 ICFA workshops organized : 2018 at SLAC, 2019 @ PSI --> Report by Andreas Adelmann (THCPL01)
    • Increasing interest of the accelerators community in these subjects
    − Machine Learning, Data Mining and Big Data Handling for Accelerators – Sandra Biedron (University of
    New Mexico) (Invited at IPAC'19, Melbourne, May 2019)
    − Machine Learning Demonstrations on Accelerators – Auralee Edelen (SLAC) (Invited at NAPAC2019,
    Lansing Michigan, September 2019
    • Not many concrete applications so far!

    13
    Data Science and Machine Learning Workshop - October 6, 2019 ICALEPCS2019 - New York, October 5-11, 2019
    Machine Learning in control systems

    14
    Data Science and Machine Learning Workshop - October 6, 2019 ICALEPCS2019 - New York, October 5-11, 2019
    ML at ICALEPCS conferences

    Keywords: Expert Systems, Machine Learning, Articicial Intelligence, Artificial Neural Networks,
    Genetic Algorithms, ...

    15
    Data Science and Machine Learning Workshop - October 6, 2019 ICALEPCS2019 - New York, October 5-11, 2019
    Past ICALEPCS tracks with ML papers

     ICALEPCS 2013 Knowledge-based Techniques

     ICALEPCS 2015 Feedback Systems, Tuning

     ICALEPCS 2017 Feedback Control and Process Tuning - Data Data


    Analytics (ML mentioned in track descriptions for the first time)

     ICALEPCS 2019 Data Analytics - Feedback Control and Process Tuning -


    Experiment Control

    16
    Data Science and Machine Learning Workshop - October 6, 2019 ICALEPCS2019 - New York, October 5-11, 2019
    ICALEPCS 2019
    ID Code Topic
    1201 MOCPL04 Tuning
    1621 WEMPR010 Anomaly Detection
    1493 MOPHA011 HMI
    1430 MOPHA043 Data Analysis
    2061 MOPHA114 Performance Optimization
    1630 MOPHA146 Tuning
    1520 TUCPL01 Prediction/Anomaly Detection/Tuning
    1594 TUCPL06  Performance Optimization
    1476 WEMPL001 Prediction
    1830 WEPHA021 Performance Optimization
    1684 WEPHA025 Fault Classification
    1667 WEPHA138  Tuning
    1464 THBPP03  Prediction/Tuning/Optimization
    1961 THCPL01  Report of 2nd ICFA workshop by Andreas Adelmann
    2021 THCPL02  Tuning
    1431 THCPL03  Performance Optimization
    1403 WEPHA123  Anomaly Detection/Virtual Diagnostics/Tuning
    2022 THCPL07  Anomaly Detection
    1543 WEPHA121  Anomaly Detection
    17
    Data Science and Machine Learning Workshop - October 6, 2019 ICALEPCS2019 - New York, October 5-11, 2019

    You might also like