Focused workshop on AI in High Energy Physics

6 Jan 2025, 09:00 → 7 Jan 2025, 23:00 Asia/Seoul

Data from particle physics is unique and challenging due to its tremendous size and complicated structure stemming  from its quantum nature.

The goal of this focus-workshop is to develop novel techniques for the future exploration of new physics, exploring exciting developments in the field of Artificial Intelligence.

The structure of the workshop is designed to survey the core questions in new physics searches and to stimulate intense discussion/collaboration among participants. 

We encourage young researchers and Ph.D students to register and to talk / discuss about their researches with experts.

Invited Speakers

• K.C. Kong (University of Kansas) 
• Sung Hak Lim (IBS-CTPU)
• Kazuki Sakurai (Warsaw University)
• David Shih (Rutgers University)
• Sangwoon Yoon (KIAS AI Center)
• Ramon Winterhalder (University of Milan)
• Andrea Wulzer (Barcelona, IFAE / ICREA, Barcelona)
• and young researchers (to be announced)

Venue: KIAS Buidling # 8,  1st floor seminar room.
For direction, please check the following website Campus Map

Notice: 

   - After this workshop, there would be KCMS-Theory Joint workshop (Jan 8 - Jan 11) at High 1 Grand Hotel. For more information, please check the following indico https://indico.kias.re.kr/e/jointworkshop2025
(The registration deadline for this Joint workshop is Dec. 20th) 

 

Monday 6 January

09:30 → 10:00 Free discussion and Coffee

09:30 → 10:00 Intro

Untitled 4.pdf

10:00 → 12:00 Session: Session 2

10:00 MadNIS - Towards the first ML event generator

Speaker: Dr Ramon Winterhalder (Università degli Studi di Milano)

madnis_kias.pdf

10:45 Generative Modeling is Imitation Learning

Speaker: Dr Sangwoong Yoon (KIAS AI Center)

20250106_swyoon_kias_hep.pdf

11:30 Free DIscussion

12:00 → 13:30 Lunch

13:30 → 14:00 Free discussion and Coffee

14:00 → 15:30 Session: Session 1

14:00 Mapping Dark Matter through the Dust of the Milky Way with Gaia DR3 and Normalizing Flows

Speaker: Prof. David Shih (Rutgers University)

Mapping Dark Matter KIAS Workshop 2025.pdf

14:45 Learning for Galactic Dynamics: Neural Stellar Density Estimation for Mapping Dark Matter in the Local Universe

Speaker: Dr Sung Hak Lim (IBS-CTPU (PTC))

KIAS_talk_short_SHLIM.pdf

15:30 → 16:00 Free discussion

16:00 → 17:00 Young Researcher Session

16:00 LeStrat-Net: Lebesgue style stratification for Monte Carlo simulations powered by machine learning

Speaker: Dr Raymundo Ramos (KIAS)

Ramos-slides.pdf

16:30 Missing information search with deep learning for mass estimation

Speaker: Dr Kayoung Ban (KIAS)

DeeLeMa_AIKIAS_2025.pdf

Tuesday 7 January

09:30 → 10:00 Free Discussion and Coffee

10:00 → 12:00 Session: Session 4

10:00 Goodness-of-fit by Neyman-Pearson Testing

Speaker: Prof. Andrea Wulzer (Barcelona, IFAE / ICREA, Barcelona)

talk_wulzer.pdf

10:45 Quantum Machine Learning

Speaker: Myeonghun Park (Seoultech)

PARK.pdf

11:30 Free discussion

12:00 → 13:30 Lunch

13:30 → 14:00 Free discussion with Coffee

14:00 → 15:30 Session: Session 5

14:00 Top Quark Polarimetry with ParticleNet

Speaker: Prof. Kyoungchul Kong (University of Kansas)

KIAS-AIHEP2025_KONG.pdf

14:45 Three-Body Entanglement in Particle Decays

Speaker: Prof. Kazuki Sakurai (University of Warsaw)

2025_01_07_KIAS_KS.pdf

15:30 → 16:00 Free discussion

16:00 → 17:00 Young Researcher Session

16:00 Detecting Light Fermiophobic Higgs Boson through CNN-based Diphoton Jet Analysis at the HL-LHC

This talk presents a novel approach to detect a light fermiophobic Higgs boson (hf) with mass in the range of 1-10 GeV at the HL-LHC, focusing on the golden channel p p → hf H → γγγγℓν. Due to the highly collimated nature of photon pairs from hf decay, traditional detection methods face significant challenges as these photons manifest as single jets rather than isolated photons. We introduce an innovative strategy utilizing Convolutional Neural Networks (CNN) to identify these distinctive diphoton jets and distinguish them from QCD backgrounds. Our method employs the Delphes framework's EFlow objects, combined with a hybrid pileup subtraction technique that merges charged hadron subtraction with SoftKiller. The analysis demonstrates remarkable results across 18 benchmark points, achieving signal significance above 5σ at an integrated luminosity of 3 ab⁻¹. For challenging scenarios involving heavy charged Higgs bosons, our CNN-based approach significantly enhances detection capability, nearly doubling the significance compared to traditional cut-based methods. This study opens new possibilities for exploring previously challenging parameter spaces in BSM physics through machine learning techniques.

Speaker: Mrs Soo Jin Lee (Konkuk University,)

KIAS-AI-HEP-diphotonjet.pdf

16:30 Can A.I. Understand Hamiltonian Mechanics?

Speaker: Tae-Geun Kim (Yonsei University)

NeuralHamilton_tgkim.pdf

18:00 → 20:00 Banquet