August 12, 2025Open Access
View Full PaperFast Jet Tagging with MLP-Mixers on FPGAs
Key Points
- Achieving state-of-the-art jet tagging accuracy, the MLP-Mixers outperform traditional methods in resource-constrained environments.
- Performance enhancements include a 97% reduction in hardware resource usage and a doubling of throughput.
- Using advanced optimization like quantization allows for real-time processing of jet constituents effectively.
- This approach sets a new benchmark for efficient machine learning applications at particle colliders, enhancing data processing capabilities.
Abstract
Abstract We explore the innovative use of MLP-Mixer models for real-time jet tagging and establish their feasibility on resource-constrained hardware like FPGAs. MLP-Mixers excel in processing sequences of jet constituents, achieving state-of-the-art performance on datasets mimicking Large Hadron Collider conditions. By using advanced optimization techniques such as High-Granularity Quantization and Distributed Arithmetic, we achieve unprecedented efficiency. These models match or surpass the accuracy of previous architectures, reduce hardware resource usage by up to 97%, double the throughput, and half the latency. Additionally, non-permutation-invariant architectures enable smart feature prioritization and efficient FPGA deployment, setting a new benchmark for machine learning in real-time data processing at particle colliders.