2 of 17

Interests

Developing use of timing information (HGTD+calo) into jet and constituent reconstruction

Related work with Brandeis group on defining vertex t0 using HGTD
Parallel work using calorimeter timing information as orthogonal t0 constraint
How to define a jet time based on PFlow objects?

Constituent and jet-level pileup suppression methods

Developed transformer-based PU jet tagging for HL-LHC (EMTopo) and studied application to per-track classification (M. Vassilev AQP)

Applying these methods for HGPFlow inputs from GPER
b-tagging at HL-LHC utilizing pileup information
Developed simple PFlow jet definition using E/p – could be established into alternative baseline to EMTopo

3 of 17

SLAC + Brandeis Group

Ariel Schwartzman (PI): XYZ % FTE
Brendon Bullard (postdoc): 5-15% FTE
Wasikul Islam (postdoc): XYZ FTE
Liangyu Wu (grad student): 10% FTE (calo t0)
Michael Cardiff (grad student): XYZ% FTE (HGTD t0)
Potential for new student(s) to join

4 of 17

Jets

GNN-JVT

Incorporate HGTD timing to JVT

Multi-vertex approach

Particle Flow:

E/P algorithm
E regression (global-jet pflow)

in-time

Overlapping jets

5 of 17

A new PFlow approach?

All approaches to PFlow start by reconstructing calorimeter clusters, and then combining them with tracks

Various methods, using ML/AI
Key challenge: removing overlaps → confusion term

Some of the confusion due to issues in topo-clustering

split/merge
overlaps with nearby particles and with pileup

Potential (new) solution:

Use tracks to guide cluster-formation!
Bring tracking from the very beginning of the process towards a more integrated calo+track approach

6 of 17

HGTD

ATLAS HL-LHC Timing is becoming a major focus. Following ATLAS Timing Workshop PC is considering creating an ATLAS Timing Forum

Keep our leadership!

Vertex t0 using HGTD

Full understanding, correlations, tails

Impact on RpT

Vertex t0 algorithm improvements
ML methods (BDT, Full regression)
Combination with LAr
Self-tagging approach!
“Bundle” tracks to improve z0 resolution

Track-HGTD matching using ML

improve purity and efficiency using ML

7 of 17

Calorimeter Timing

ML Vertex t0 reconstruction
Data studies
Crosstalk measurement
Intercalibration of LAr/HGTD
…

8 of 17

4D Tracking

CERN group started to work on the ITk replacement!

ATLAS expected to launch official effort next year

Layouts

Double second layer
Implement/study full simulation layouts and their performance

Expand physics case
6D Tracking

Study impact on seeding and track reconstruction
Application to muon and pp collider

Vertex Selection at FCC-hh

4D vertexing vs jet-vertex association

Cutting-edge hardware opportunities

First 3D Integrated LGAD sensors and electronics!
4D Tracking demonstrator (Telescope)
SPADs for tracking

9 of 17

b-tagging

Global b-tagging approach in ATLAS
Add PU vertices
Should lead to ATLAS PUB note / paper and next LHC default

If successful, this would be a major accomplishment for the whole LHC and future colliders

Other applications of global-tagging?

10 of 17

5D Dual Readout Calorimetry

Benchmark jet resolution characterization
Digital SiPMs and prototype for fiber HAD IDEA Calorimeter
5D Particle Flow

C+S+t+tracks ML reconstruction

Crystal EM calorimeter with timing
5D Calorimeter performance for PID and LLP
Interplay between 5D Calo and Timing Layers

11 of 17

Bonus: XFEL Photon Collider

e^-

ɣ₀: polarized

laser beam

Compton IP

(IPC)

e^-ɣ₀ → e^-ɣ

ɣ: Compton gamma-ray

photons

e^-

ɣɣ

Primary IP

ɣɣ collisions

e^-ɣ collisions

e^-e^- collisions

e⁺e^- collisions

Compton IP

(IPC)

E(ɣ)~ E₀

E(e^-)=E₀

ɣɣ₀ → e⁺e^-

Lɣɣ = 𝜿² L_e-e-

𝜿²= Compton efficiency

A Photon-Photon Collider is also a e^-ɣ, e^-e^-, and e⁺e^- collider!

https://docs.google.com/presentation/d/1qotjDDzNQ3FTETYOWMqqcs0hplbAG2DQ/edit?usp=sharing&ouid=115946389821264344062&rtpof=true&sd=true

12 of 17

Photon Collider Concepts

Optical photon colliders concepts were first proposed in the 1980s and 1990s and developed in the 2000’s mainly as a complementary second interaction region of e⁺e^- linear colliders

Produce similar number of Higgs bosons per year than e⁺e^-, but with higher backgrounds

Recent innovation in photon science, particularly in XFELs lead to enhanced capabilities

XCC

SAPPHiRE

CLICHE

13 of 17

XCC Single Higgs Sensitivity

Extremely precise determination of Cross Sections times Branching Ratios

14 of 17

Higgs Self Coupling

Δk_λ calculated assuming 11% cross-section uncertainty for HH→4b at 280 GeV CoM and assuming the same sensitivity to all HH decay topologies

HL-LHC

XCC

FCC-hh

LCF 550 GeV

LCF + XCC

15 of 17