paper-with-me

홈 › Papers

Low Latency Transformer Inference on FPGAs for Physics Applications with hls4ml

2024-09-08 · Zhixing Jiang, Dennis Yin, Yihui Chen, Elham E Khoda, Scott Hauck, Shih-Chieh Hsu, Ekaterina Govorkova, Philip Harris, Vladimir Loncar, Eric A. Moreno

This study presents an efficient implementation of transformer architectures in Field-Programmable Gate Arrays(FPGAs) using hls4ml. We demonstrate the strategy for implementing the multi-head attention, softmax, and normalization layer and evaluate three distinct models. Their deployment on VU13P FPGA chip achieved latency less than 2us, demonstrating the potential for real-time applications. HLS4ML compatibility with any TensorFlow-built transformer model further enhances the scalability and applicability of this work. Index Terms: FPGAs, machine learning, transformers, high energy physics, LIGO

📄 PDF Abstract BibTeX arXiv:2409.05207

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

Fast inference of deep neural networks in FPGAs for particle physics

2018-04-16 · Javier Duarte, Song Han, Philip Harris, Sergo Jindariani 외

Recent results at the Large Hadron Collider (LHC) have pointed to enhanced physics capabilities through the improvement of the real-time event processing techniques. Machine learning methods are ubiquitous and have prove…

BIG-bench Machine LearningHigh-Level Synthesis

Sub-microsecond Transformers for Jet Tagging on FPGAs

2025-10-26 · Lauri Laatu, Chang Sun, Arianna Cox, Abhijith Gandrakota 외 arxiv

We present the first sub-microsecond transformer implementation on an FPGA achieving competitive performance for state-of-the-art high-energy physics benchmarks. Transformers have shown exceptional performance on multipl…

Jet Tagging

Ultra-low latency recurrent neural network inference on FPGAs for physics applications with hls4ml

2022-07-01 · Elham E Khoda, Dylan Rankin, Rafael Teixeira de Lima, Philip Harris 외

Recurrent neural networks have been shown to be effective architectures for many tasks in high energy physics, and thus have been widely adopted. Their use in low-latency environments has, however, been limited as a resu…

Fast inference of Boosted Decision Trees in FPGAs for particle physics

2020-02-05 · Sioni Summers, Giuseppe Di Guglielmo, Javier Duarte, Philip Harris 외

We describe the implementation of Boosted Decision Trees in the hls4ml library, which allows the translation of a trained model into FPGA firmware through an automated conversion process. Thanks to its fully on-chip impl…

Translation

Position: The Need for Ultrafast Training

2026-02-02 · Duc Hoang arxiv

Domain-specialized FPGAs have delivered unprecedented performance for low-latency inference across scientific and industrial workloads, yet nearly all existing accelerators assume static models trained offline, relegatin…