"ML at HEP workshop" in Japan 2024

9 Jan 2024, 12:30 → 10 Jan 2024, 17:30 Asia/Tokyo

Kobayashi hall (KEK (High Energy Accelerator Research Organization))

We are delighted to announce that our second annual "Machine Learning at HEP workshop" will take place between 9th-10th January 2024 at High Energy Accelerator Research Organization (KEK) in Japan.

In the previous workshop ( KEK indico link ) , we successfully overviewed the state-of-the-art Machine Learning (ML) applications to high energy physics, and had lively discussions about the directions of ML developments in the field. The scope of this workshop is to highlight recent progresses in the state-of-the-art ML applications to high energy physics and astrophysics through several talks by world-leading researchers, and to overview the related researches progressing in Japan.

The workshop format is hybrid (in-person at KEK (strongly preferred for lively discussions) and online via zoom (possible)). Long review presentations are given by invited speakers, while a session of several contribution talks is also scheduled. Please indicate your intention to present a contribution talk on your ML-related work via the registration form by December 15th. Domestic travel and accommodation support can be provided to a limited number of presenters. If you would like to request it, please contact us via the LOC mailing list by December 15th.

Registration: 

Registration for participation was open at https://conference-indico.kek.jp/event/253/registrations/ , and  has been closed on December 25th. No registration fee is necessary. Application for contribution talks has been closed.

Confirmed Invited Speakers:

• Matt Buckley (Rutgers)
• Gregor Kasieczka (Hamburg)
• Shiro Ikeda (The Institute of Statistical Mathematics)
• François Lanusse (Flatiron Institute / CNRS)
• Sung Hak Lim (Rutgers) 
• Vinicius Mikuni (LBL)
• Mariel Pettee (LBL)
• Huilin Qu (CERN)

---

This workshop will be hosted co-organized by KEK WPI-QUP (International Center for Quantum-field Measurement Systems for Studies of the Universe and Particles) and IPNS (Institute of Particle and Nuclear Studies), and supported by KEK WPI-QUP, JSPS Core-to-Core Program "AI-Smart" and JSPS Grant-in-Aid for Transformative Research Areas (A) "Machine Learning Physics".

Program organizers: Yuji Chinone (KEK), Hironao Miyatake (Nagoya U), Yu Nakahama (KEK), Mihoko Nojiri (KEK), Junichi Tanaka (U Tokyo)

Local organisers: Yuji Chinone (KEK), Daniel Jeans (KEK), Gaku Mitsuka (KEK), Kunihiro Nagano (KEK), Kodai Matsuoka (KEK), Yu Nakahama (KEK), Mihoko Nojiri (KEK)

Group_photo.JPG

Tuesday, 9 January

12:30 Registration

1 Opening

Speaker: Yu NAKAHAMA (KEK)

Welcome.pdf

2 Data Science for Astronomy

Speaker: Shiro Ikeda (The Institute of Statistical Mathematics)

2024.0109.KEK_ikeda.pdf

3 Review: Low-level object reconstruction and simulation using ML

Speaker: Vinicius Mikuni (LBL)

Ml@HEP_Reco.pptx

4 ML for Jet Tagging: Status and Prospects

Speaker: Huilin Qu (CERN)

JetTaggingML_H_Qu_20240109.pdf

15:05 Group Photo

15:10 Coffee

5 Dark Matter Applications of Machine Learning

Speaker: Matt Buckley (Rutgers University)

KEK2024_buckley.pdf

Abstract: Machine learning algorithms provide powerful new windows into complicated and high-dimensional datasets. Of particular interests are normalizing flows, which estimate the phase space density of data in an unsupervised manner. These new techniques arrive concurrently with an era of Big Data in astrophysics, as large sky surveys provide a wealth of data about the structure of our own Galaxy and the Universe beyond. In this talk, I discuss applications of normalizing flows to Gaia data, which measures the position and velocity of the nearest 1.5 billion bright stars (30 million with full 6D kinematics): mapping the dark matter distribution within 4 kpc of the Sun, searching for stellar streams in the Galaxy, and generating synthetic astronomical datasets.

6 Flow-Matching models for the LHC

Speaker: Darius Faroughy (Rutgers University)

ML_farougy.pdf

7 AI Methods in Galaxy Formation

Speaker: Suchetha Cooray (NAOJ)

8 Multi-Scale Cross-Attention Transformer Encoder for Event Classification

Speaker: Ahmed Hammad (KEK)

Talk_Hammad.pdf

9 A data-driven and model-agnostic approach to solving combinatorial assignment problems in searches for new physics

Speaker: Javier Montejo (IFAE Barcelona & QUP)

20240109_MLHEP_unsupervised.pdf

Wednesday, 10 January

10 Prospects of LLMs for Fundamental Physics

Speaker: Mariel Pettee (LBL)

Prospects of LLMs for Fundamental Physics.pptx

11 Review and Perspective on AI for Observational Cosmology

Speaker: François Lanusse (Simons Foundation/CNRS)

Link

12 Extracting optimal information from upcoming cosmological surveys

Speaker: Adrian Bayer (Princeton)

13 Cosmosage: a natural-language assistant for cosmology

Speaker: Tijmen de Haan (KEK)

TdH_MLatHEP_cosmosage.pdf

11:30 Lunch

14 Using AI for unsupervised discovery in particle physics

Speaker: Gregor Kasieczka (Hamburg)

DeepLearning_Japan_KEK_2024.pdf

15 Model-independence of machine-learning-based mass reconstruction for inclusive multi-jet search

Speaker: Takane Sano (Kyoto University)

tsano_slides_model-indep_v2.pdf

16 Jet Classification using High-Level Features from Anatomy of Top Jets

Speaker: Sung Hak Lim (Rutgers University)

MLHEP.pdf

17 Development of Anomaly Detection techniques applied to the building and Quality Control of ATLAS new silicon tracking detector

Speaker: Louis Vaslin (KEK QUP)

VASLIN_slides_2024-01-10.pdf

18 Pre-training strategy using real particle collision data for event classification in collider physics

Speaker: Tomoe Kishimoto (KEK)

2024_01_10_tomoe.pdf

15:00 Coffee break

19 Returning CP-observables to the frames they belong (unfolding)

Speaker: Rahool Barman (IPMU)

ML@HEP_Unfolding.pdf

20 A Machine-learning Approach to Assessing the Presence of Substructure in Quasar-host Galaxies

Speaker: Chris Nagele (University of Tokyo)

KEK.pdf

21 Improving the diphoton event selection with GNNs

Speaker: Isai Roberto Sotarriva Alvarez (TokyoTech)

MLColloquium2024_v3.pdf

22 Machine-learning-assisted beam tuning at the KEK Linac and prospects for SuperKEKB

Speaker: Shinnosuke Kato (University of Tokyo)

ML_at_HEP2024ver3.pdf

23 Development of a hardware trigger using machine learning in the Belle II experiment

Speaker: Riku Nomaru (University of Tokyo)

ml_at_hep_nomaru.pdf

24 Closing

Speaker: Junichi Tanaka (The University of Tokyo )

Closing_JT_20240110.pdf