mlsanityData to 14 Oct 2026

Open-source FPGA-ML codesign for the MLPerf Tiny Benchmark

arXiv 2206.11791 ↗First posted 23 Jun 2022cs.LG cs.ARcited by 9

Not on the map. Why?

Abstract (arXiv)

We present our development experience and recent results for the MLPerf Tiny Inference Benchmark on field-programmable gate array (FPGA) platforms. We use the open-source hls4ml and FINN workflows, which aim to democratize AI-hardware codesign of optimized neural networks on FPGAs. We present the design and implementation process for the keyword spotting, anomaly detection, and image classification benchmark tasks. The resulting hardware implementations are quantized, configurable, spatial dataflow architectures tailored for speed and efficiency and introduce new generic optimizations and common workflows developed as a part of this work. The full workflow is presented from quantization-aware training to FPGA implementation. The solutions are deployed on system-on-chip (Pynq-Z2) and pure FPGA (Arty A7-100T) platforms. The resulting submissions achieve latencies as low as 20 μs and energy consumption as low as 30 μJ per inference. We demonstrate how emerging ML benchmarks on heterogeneous hardware platforms can catalyze collaboration and the development of new techniques and more accessible tools.

Key facts. Cited by 9 papers in the map (data through 14 Oct 2026). Builds on 31 of its 58 references that are in the map. First posted on arXiv 23 Jun 2022.

Builds on

31 of 58 references.

References · 31

  1. 2015Deep Residual Learning for Image Recognitioncited by 52,096
  2. 2019PyTorch: An Imperative Style, High-Performance Deep Learning Librarycited by 20,305
  3. 2014Very Deep Convolutional Networks for Large-Scale Image Recognitioncited by 15,364
  4. 2018Speech Commands: A Dataset for Limited-Vocabulary Speech Recognitioncited by 752
  5. 2018A System for Massively Parallel Hyperparameter Tuningcited by 202
  6. 2016FINN: A Framework for Fast, Scalable Binarized Neural Network Inferencecited by 150
  7. 2018Fast inference of deep neural networks in FPGAs for particle physicscited by 133
  8. 2021MLPerf Tiny Benchmarkcited by 124
  9. 2017Hello Edge: Keyword Spotting on Microcontrollerscited by 110
  10. 2020Benchmarking TinyML Systems: Challenges and Directioncited by 91
  11. 2021hls4ml: An Open-Source Codesign Workflow to Empower Scientific Low-Power Machine Learning Devicescited by 62
  12. 2018FINN-R: An End-to-End Deep-Learning Framework for Fast Exploration of Quantized Neural Networkscited by 56
19 more
  1. 2019ToyADMOS: A Dataset of Miniature-Machine Operating Sounds for Anomalous Sound Detectioncited by 44
  2. 2020Compressing deep neural networks on FPGAs to binary and ternary precision with HLS4MLcited by 42
  3. 2021Fast convolutional neural networks on FPGAs with hls4mlcited by 42
  4. 2020Distance-Weighted Graph Neural Networks on FPGAs for Real-Time Particle Reconstruction in High Energy Physicscited by 34
  5. 2021Applications and Techniques for Fast Machine Learning in Sciencecited by 32
  6. 2021Ps and Qs: Quantization-aware pruning for efficient low latency neural network inferencecited by 31
  7. 2019FPGA-based Accelerators of Deep Learning Networks for Learning and Classification: A Reviewcited by 29
  8. 2018Toolflows for Mapping Convolutional Neural Networks on FPGAs: A Survey and Future Directionscited by 27
  9. 2020Mix and Match: A Novel FPGA-Centric Deep Neural Network Quantization Frameworkcited by 27
  10. 2020Fast inference of Boosted Decision Trees in FPGAs for particle physicscited by 26
  11. 2021Graph Neural Networks for Charged Particle Tracking on FPGAscited by 17
  12. 2018Accelerating CNN inference on FPGAs: A Surveycited by 16
  13. 2017Streamlined Deployment for Quantized Neural Networkscited by 12
  14. 2019FixyNN: Efficient Hardware for Mobile Computer Vision via Transfer Learningcited by 10
  15. 2022QONNX: Representing Arbitrary-Precision Quantized Neural Networkscited by 10
  16. 2016Caffeinated FPGAs: FPGA Framework For Convolutional Neural Networkscited by 8
  17. 2017fpgaConvNet: A Toolflow for Mapping Diverse Convolutional Neural Networks on Embedded FPGAscited by 5
  18. 2018Quantized Guided Pruning for Efficient Hardware Implementations of Convolutional Neural Networkscited by 5
  19. 2019A flexible FPGA accelerator for convolutional neural networkscited by 3

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2022: 1; 2023: 1; 2024: 4; 2025: 2; 2026: 1.

9 papers.

  1. 2025hls4ml: A Flexible, Open-Source Platform for Deep Learning Acceleration on Reconfigurable Hardwarecited by 9
  2. 2023Tailor: Altering Skip Connections for Resource-Efficient Inferencecited by 6
  3. 2024rule4ml: An Open-Source Tool for Resource Utilization and Latency Estimation for ML Models on FPGAcited by 3
  4. 2024Architectural Implications of Neural Network Inference for High Data-Rate, Low-Latency Scientific Applicationscited by 2
  5. 2024CGRA4ML: A Hardware/Software Framework to Implement Neural Networks for Scientific Edge Computingcited by 2
  6. 2025wa-hls4ml: A Benchmark and Surrogate Models for hls4ml Resource and Latency Estimationcited by 2
  7. 2024PEFSL: A deployment Pipeline for Embedded Few-Shot Learning on a FPGA SoCcited by 1
  8. 2026BitLogic: Training Framework for Gradient-Based FPGA-Native Neural Networkscited by 1
  9. 2022Neural network accelerator for quantum controlcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2024A survey on FPGA-based accelerator for ML models
  2. 2026Energy Efficient Software Hardware CoDesign for Machine Learning: From TinyML to Large Language Models
  3. 2025Energy-Efficient FPGA Framework for Non-Quantized Convolutional Neural Networks
  4. 2025Neural Network Quantization for Microcontrollers: A Comprehensive Survey of Methods, Platforms, and Applications
  5. 2025SIRA: Scaled-Integer Range Analysis for Optimizing FPGA Dataflow Neural Network Accelerators
  6. 2025FPGA Co-Design for Efficient N:M Sparse and Quantized Model Inference
  7. 2023Shedding the Bits: Pushing the Boundaries of Quantization with Minifloats on FPGAs
  8. 2026VQ-LIC: Shared Vector-Quantized Learned Image Compression on a Resource-Constrained FPGA ↗