UHH Trigger & FastML Meeting

→ Europe/Berlin
SR 125 (Bdg. 68)

SR 125

Bdg. 68
Description

zoom link

Hide

Introduction

New meeting series, mostly to report on Trigger activities across the UHH groups.

Timeslot was decided in advance via poll, this was the only viable option - can explore alternatives if absolutely necessary (e.g. availabilities shift drastically).

Plan to mostly focus on roundtable, with occasional long(er) form presentations.

Have added roundtable to indico event with individual subcontributions, please contact Carsten if you want a subcontribution to be added for yourself to upload and discuss material.

Please feel free to invite additional people as you see fit.

Roundtable Reports

Robert Ward

  • Working on L1 menu readiness for the HL-LHC.
  • Benchmarking and DQM studies ongoing.

Matthias Schröder

  • Project focus: di-Higgs topology trigger menu.

Karim El-Morabit

  • Roles: HLT stream convener; former scouting convener.
  • Project: integration of anomaly triggers.
  • Supervision: Sven working on L1 anomaly detection (initially AXOL1TL implementation, now focused on understanding data).

Matteo Bonanomi

  • Role: Muon L2 convenership.
  • Project: Phase-2 tools for L1 triggers, with partial use already in Run 3 (institutional commitment).

Carsten Burgard

  • Project: Monitoring of ML triggers to track changes with run conditions and study need for retraining.

Gregor Kasieczka

  • Proposal: define Master’s thesis topics—ideally two: one CMS-related, one ML-focused.
  • Suggestion: create a project poster for teaching rooms (contact: Anna).
  • Opening a PhD position on long-lived particles; part of L1 trigger service task.
  • QU Smart 2.3 ML project carried by via FastML Lab; relevant researchers across campus should exchange updates at QU Days.

Any Other Business

Discussion of scope & purpose of the meeting

  • Carsten: Meeting intended primarily as a forum for students to report on trigger-related work.
  • Matteo: Students often report only in the Higgs meeting; useful to define this meeting as a dedicated trigger/fast-ML reporting space.
  • Gregor: Students should continue presenting Higgs-related work in the Higgs meeting, but report trigger-relevant items here.
  • Matteo: Agreed—this provides a dedicated forum outside the Higgs meeting, which many students attend without working on trigger topics.

Mattermost

  • There is a UHH Mattermost group "ML@L1", please join!
There are minutes attached to this event. Show them.