# Learning Before Filtering: Real-Time Hardware Learning at the Detector Level

arXiv: https://arxiv.org/abs/2506.11981 · HTML: https://mlsanity.com/paper/2506.11981/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 11 of its 40 references that are in the map.
- First posted on arXiv 13 Jun 2025.

## Builds on (most cited first, 11 of 11)

- [Delving Deep into Rectifiers: Surpassing Human-Level Performance on ImageNet Classification](https://mlsanity.com/paper/1502.01852/) (2015)
- [Fast inference of deep neural networks in FPGAs for particle physics](https://mlsanity.com/paper/1804.06913/) (2018)
- [ParticleNet: Jet Tagging via Particle Clouds](https://mlsanity.com/paper/1902.08570/) (2019)
- [Compute-Efficient Deep Learning: Algorithmic Trends and Opportunities](https://mlsanity.com/paper/2210.06640/) (2022)
- [Distance-Weighted Graph Neural Networks on FPGAs for Real-Time Particle Reconstruction in High Energy Physics](https://mlsanity.com/paper/2008.03601/) (2020)
- [Applications and Techniques for Fast Machine Learning in Science](https://mlsanity.com/paper/2110.13041/) (2021)
- [FastML Science Benchmarks: Accelerating Real-Time Scientific Edge Machine Learning](https://mlsanity.com/paper/2207.07958/) (2022)
- [Benchmarking Deep Learning Models on NVIDIA Jetson Nano for Real-Time Systems: An Empirical Investigation](https://mlsanity.com/paper/2406.17749/) (2024)
- [A survey on FPGA-based accelerator for ML models](https://mlsanity.com/paper/2412.15666/) (2024)
- [Innovations in trigger and data acquisition systems for next-generation physics facilities](https://mlsanity.com/paper/2203.07620/) (2022)
- [Machine Learning Pipelines with Modern Big Data Tools for High Energy Physics](https://mlsanity.com/paper/1909.10389/) (2019)

## Abstract (arXiv)

Advances in sensor technology and automation have ushered in an era of data abundance, where the ability to identify and extract relevant information in real time has become increasingly critical. Traditional filtering approaches, which depend on a priori knowledge, often struggle to adapt to dynamic or unanticipated data features. Machine learning offers a compelling alternative-particularly when training can occur directly at or near the detector. This paper presents a digital hardware architecture designed for real-time neural network training, specifically optimized for high-throughput data ingestion. The design is described in an implementation-independent manner, with detailed analysis of each architectural component and their performance implications. Through system parameterization, the study explores trade-offs between processing speed, model complexity, and hardware resource utilization. Practical examples illustrate how these parameters affect applicability across various use cases. A proof-of-concept implementation on an FPGA demonstrates in-situ training, confirming that computational accuracy is preserved relative to conventional software-based approaches. Moreover, resource estimates indicate that current-generation FPGAs can train networks of approximately 3,500 neurons per chip. The architecture is both scalable and adaptable, representing a significant advancement toward integrating learning directly within detector systems and enabling a new class of extreme-edge, real-time information processing.