Enabling Low-Latency Machine learning on Radiation-Hard FPGAs with hls4ml

TL;DR

This paper demonstrates a low-latency, radiation-hard machine learning application on FPGAs, introducing a new backend for hls4ml to support the PolarFire FPGA, enabling efficient on-detector ML in high-energy physics experiments.

Contribution

We developed a radiation-hard FPGA backend for hls4ml, enabling automatic translation of ML models into HLS projects for PolarFire FPGAs, and demonstrated a low-latency autoencoder for high-energy physics applications.

Findings

Autoencoder achieves 25 ns latency on PolarFire FPGA.

Model quantized to 10-bit weights with minimal performance loss.

Resources are low enough for in-FPGA deployment within protected logic.

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 PolarFire family of FPGAs, one of the few commercially available and radiation hard FPGAs. We present the synthesis of the autoencoder on a target PolarFire FPGA, which indicates that a latency of 25 ns can be achieved. We show that the resources utilized are low enough that the model can be placed within the inherently protected logic of the FPGA. Our extension to hls4ml is a significant contribution, paving the way for broader adoption of ML on FPGAs in high-radiation environments.