LL-GNN: Low Latency Graph Neural Networks on FPGAs for Particle Detectors

@article{Que2022LLGNNLL,
  title={LL-GNN: Low Latency Graph Neural Networks on FPGAs for Particle Detectors},
  author={Zhiqiang Que and Marcus Loo and Hongxiang Fan and Michaela Blott and Maurizio Pierini and Alexander Tapper and Wayne Luk},
  journal={ArXiv},
  year={2022},
  volume={abs/2209.14065}
}
—This work proposes a novel reconfigurable architecture for low latency Graph Neural Network (GNN) design specifically for particle detectors. Adopting FPGA-based GNNs for particle detectors is challenging since it requires sub-microsecond latency to deploy the networks for online event selection in the Level-1 triggers for the CERN Large Hadron Collider experiments. This paper proposes a custom code transformation with strength reduction for the matrix multiplication operations in the… 
1 Citations

Accelerating Transformer Neural Networks on FPGAs for High Energy Physics Experiments

A novel TNN-based architecture, efficiently mapped to Field-Programmable Gate Arrays, that outperforms GPU inference capabilities involving state-of-the-art neural network models by approximately 1000 times while preserving comparable classification accuracy is proposed.

References

SHOWING 1-10 OF 54 REFERENCES

JEDI-net: a jet identification algorithm based on interaction networks

We investigate the performance of a jet identification algorithm based on interaction networks (JEDI-net) to identify all-hadronic decays of high-momentum heavy particles produced at the LHC and

Graph Neural Networks for Charged Particle Tracking on FPGAs

An automated translation workflow is introduced, integrated into a broader tool called hls4ml, for converting GNNs into firmware for field-programmable gate arrays (FPGAs), which could enable the inclusion of charged particle tracking Gnns at the trigger level for HL-LHC experiments.

The Phase-2 Upgrade of the CMS Level-1 Trigger

This Technical Design Report describes the ongoing developments and plans towards the upgrade of the CMS Level-1 trigger for the High-Luminosity Large Hadron Collider.

DNNExplorer: A Framework for Modeling and Exploring a Novel Paradigm of FPGA-based DNN Accelerator

A novel FPGA-based DNN accelerator design paradigm and its automation tool, called DNNExplorer, are proposed to enable fast exploration of various accelerator designs under the proposed paradigm and deliver optimized accelerator architectures for existing and emerging DNN networks.

Unrolling Ternary Neural Networks

This article demonstrates how to remove 90% of the operations in convolutional layers by exploiting sparsity and compile-time optimizations in the datapath and routing of an FPGA.

Reconfigurable Acceleration of Graph Neural Networks for Jet Identification in Particle Physics

  • Zhiqiang QueMarcus LooW. Luk
  • Computer Science
    2022 IEEE 4th International Conference on Artificial Intelligence Circuits and Systems (AICAS)
  • 2022
A novel reconfigurable architecture to accelerate Graph Neural Networks (GNNs) for JEDI-net, a jet identification algorithm in particle physics which achieves state-of-the-art accuracy, is presented, which avoids the costly multiplication of the adjacency matrix with the input feature matrix.

Automatic heterogeneous quantization of deep neural networks for low-latency inference on the edge for particle detectors

A method for designing optimally heterogeneously quantized versions of deep neural network models for minimum-energy, high-accuracy, nanosecond inference and fully automated deployment on chip is introduced.

EGCN: An Efficient GCN Accelerator for Minimizing Off-Chip Memory Access

An efficient GCN inference accelerator, EGCN, specialized for minimizing off- chip memory access is introduced, which achieves 41.9% off-chip DRAM access reduction, 1.49× speedup, and 1.95× energy efficiency improvement on average over the state-of-the-art accelerators.

Adaptable Butterfly Accelerator for Attention-based NNs via Hardware and Algorithm Co-design

FABNet is proposed, a hardware-friendly variant that adopts a unified butterfly sparsity pattern to approximate both the attention mechanism and the FFNs and a novel adaptable butterfly accelerator is proposed that can be configured at runtime via dedicated hardware control to accelerate different butterfly layers using a single unified hardware engine.

DRGN: a dynamically reconfigurable accelerator for graph neural networks

...