![[2602.15751] Enabling Low-Latency Machine learning on Radiation-Hard FPGAs with hls4ml](https://arxiv.org/static/browse/0.3.4/images/arxiv-logo-fb.png){kind=logo}
[2602.15751] Enabling Low-Latency Machine learning on Radiation-Hard FPGAs with hls4ml
Summary
This article presents a novel approach to implementing low-latency machine learning on radiation-hard FPGAs, demonstrating its application in high-energy physics experiments.
Why It Matters
The integration of machine learning with radiation-hard FPGAs is crucial for advancing high-energy physics experiments, where traditional ML tools often fail due to environmental constraints. This research paves the way for more efficient data processing in challenging conditions, potentially enhancing experimental outcomes.
Key Takeaways
- Demonstrates a lightweight autoencoder for compressing timing readouts in high-energy physics.
- Introduces a hardware-aware quantization strategy, achieving 10-bit weights with minimal performance loss.
- Develops a new backend for hls4ml, enabling automatic translation of ML models for radiation-hard FPGAs.
High Energy Physics - Experiment arXiv:2602.15751 (hep-ex) [Submitted on 17 Feb 2026] Title:Enabling Low-Latency Machine learning on Radiation-Hard FPGAs with hls4ml Authors:Katya Govorkova, Julian Garcia Pardinas, Vladimir Loncar, Victoria Nguyen, Sebastian Schmitt, Marco Pizzichemi, Loris Martinazzoli, Eluned Anne Smith View a PDF of the paper titled Enabling Low-Latency Machine learning on Radiation-Hard FPGAs with hls4ml, by Katya Govorkova and 7 other authors View PDF HTML (experimental) Abstract:This paper presents the first demonstration of a viable, ultra-fast, radiation-hard machine learning (ML) application on FPGAs, which could be used in future high-energy physics experiments. We present a three-fold contribution, with the PicoCal calorimeter, planned for the LHCb Upgrade II experiment, used as a test case. First, we develop a lightweight autoencoder to compress a 32-sample timing readout, representative of that of the PicoCal, into a two-dimensional latent space. Second, we introduce a systematic, hardware-aware quantization strategy and show that the model can be reduced to 10-bit weights with minimal performance loss. Third, as a barrier to the adoption of on-detector ML is the lack of support for radiation-hard FPGAs in the High-Energy Physics community's standard ML synthesis tool, hls4ml, we develop a new backend for this library. This new back-end enables the automatic translation of ML models into High-Level Synthesis (HLS) projects for the Microchip Po...
