NGT workshop: hls4ml HEP Community Forum

Contribution List

2. The NGT WP1.2 and WP1.3

Michael Kagan (SLAC National Accelerator Laboratory (US))

27/09/2024, 09:00

1. The hls4ml project: status and perspectives

Vladimir Loncar (Massachusetts Inst. of Technology (US))

27/09/2024, 09:30

8. ATLAS 2.1: L0 Global

Ioannis Xiotidis (CERN)

27/09/2024, 10:30

User contributions

9. ATLAS 2.2: L0 Muon

Rimsky Alejandro Rojas Caballero (University of Massachusetts (US))

27/09/2024, 11:00

User contributions

10. ATLAS 2.4: Event Filter Tracking

Julian Wollrath (CERN)

27/09/2024, 11:30

User contributions

11. CMS 3.5: L1 scouting

Giovanni Petrucciani (CERN)

27/09/2024, 13:30

User contributions

12. CMS 3.6: ML objects in L1 trigger

Sioni Paris Summers (CERN)

27/09/2024, 14:00

User contributions

13. CMS 3.7: Compression/Anomaly detection

Jennifer Ngadiuba (FNAL), Maciej Mikolaj Glowacki (CERN)

27/09/2024, 14:30

User contributions

14. Online track reconstruction with graph neural networks on FPGAs for the ATLAS experiment

Santosh Parajuli (Univ. Illinois at Urbana Champaign (US)), santosh parajuli

27/09/2024, 15:30

User contributions

15. Smart pixels with data reduction at the source

Jannicke Pearkes (University of Colorado Boulder (US))

27/09/2024, 15:45

User contributions

17. Machine learning evaluation in the Global Event Processor FPGA for the ATLAS trigger upgrade

Ben Carlson (Westmont College)

27/09/2024, 16:00

User contributions

18. hls4ml web interface

Christopher Edward Brown (CERN)

27/09/2024, 16:15

User contributions

19. FastML general meeting

27/09/2024, 17:00

Discussion on the formation of FastML Foundation

6. hls4ml web interface

Christopher Edward Brown (CERN)

Short talk (5min +3min questions)

Quickly understanding the resource usage of a model is a common workflow for anyone designing an FPGA targeted model. Often the developer won't have access to vivado or a PC capable of running it. A website where the user can upload a trained model, hls4ml can run on the model, produce firmware estimates and a package of the associated firmware could solve this issue. A web interface GUI could...

4. Machine learning evaluation in the Global Event Processor FPGA for the ATLAS trigger upgrade

Ben Carlson (Westmont College), Santosh Parajuli (Univ. Illinois at Urbana Champaign (US))

Short talk (5min +3min questions)

The Global Event Processor (GEP) FPGA is an area-constrained, performance-critical element of the Large Hadron Collider's (LHC) ATLAS experiment. It needs to very quickly determine which small fraction of detected events should be retained for further processing, and which other events will be discarded. This system involves a large number of individual processing tasks, brought together...

3. Online track reconstruction with graph neural networks on FPGAs for the ATLAS experiment

Santosh Parajuli (Univ. Illinois at Urbana Champaign (US))

Long talk (10min +5min questions)

For the HL-LHC upgrade of the ATLAS TDAQ system, a heterogeneous computing farm deploying GPUs and/or FPGAs is under study, together with the use of modern machine learning algorithms such as Graph Neural Networks (GNNs). We present a study on the reconstruction of tracks in the ATLAS Inner Tracker using GNNs on FPGAs for the Event Filter system. We explore each of the steps in a GNN-based...

5. Reusability in emulation and firmware

Sioni Paris Summers (CERN)

Long talk (10min +5min questions)

With hls4ml now deployed online at CMS, we may look back on the deployment experience with a view to improvements for the years ahead. In particular, the compile-time fixing of network architecture and parameters has some implications. In emulation, model updates are tied to CMS Software release schedules and cannot be updated on changing conditions. Likewise in firmware, models are baked into...

16. Reusability in emulation and firmware

Sioni Paris Summers (CERN)

User contributions

7. Smart pixels with data reduction at the source

Jannicke Pearkes (University of Colorado Boulder (US))

Long talk (10min +5min questions)

Highly granular pixel detectors allow for increasingly precise measurements of charged particle tracks. At the High Luminosity Large Hadron Collider, data rates from the pixel detector exceed those feasible for integration in the Level 1 trigger. However, the shape of charge clusters deposited in the pixel sensors can be used to determine the physical properties of the traversing particle,...