mlsanityData to 7 Oct 2026

DORY: Automatic End-to-End Deployment of Real-World DNNs on Low-Cost IoT MCUs

arXiv 2008.07127 ↗First posted 17 Aug 2020cs.DC cs.AR cs.NEcited by 57

Not on the map. Why?

Abstract (arXiv)

The deployment of Deep Neural Networks (DNNs) on end-nodes at the extreme edge of the Internet-of-Things is a critical enabler to support pervasive Deep Learning-enhanced applications. Low-Cost MCU-based end-nodes have limited on-chip memory and often replace caches with scratchpads, to reduce area overheads and increase energy efficiency -- requiring explicit DMA-based memory transfers between different levels of the memory hierarchy. Mapping modern DNNs on these systems requires aggressive topology-dependent tiling and double-buffering. In this work, we propose DORY (Deployment Oriented to memoRY) - an automatic tool to deploy DNNs on low cost MCUs with typically less than 1MB of on-chip SRAM memory. DORY abstracts tiling as a Constraint Programming (CP) problem: it maximizes L1 memory utilization under the topological constraints imposed by each DNN layer. Then, it generates ANSI C code to orchestrate off- and on-chip transfers and computation phases. Furthermore, to maximize speed, DORY augments the CP formulation with heuristics promoting performance-effective tile sizes. As a case study for DORY, we target GreenWaves Technologies GAP8, one of the most advanced parallel ultra-low power MCU-class devices on the market. On this device, DORY achieves up to 2.5x better MAC/cycle than the GreenWaves proprietary software solution and 18.1x better than the state-of-the-art result on an STM32-F746 MCU on single layers. Using our tool, GAP-8 can perform end-to-end inference of a 1.0-MobileNet-128 network consuming just 63 pJ/MAC on average @ 4.3 fps - 15.4x better than an STM32-F746. We release all our developments - the DORY framework, the optimized backend kernels, and the related heuristics - as open-source software.

Key facts. Cited by 57 papers in the map (data through 7 Oct 2026). First posted on arXiv 17 Aug 2020.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 4; 2022: 8; 2023: 14; 2024: 16; 2025: 8; 2026: 7.

57 papers.

Sort:

  1. 2022On-Device Training Under 256KB Memorycited by 158
  2. 2024Tiny Machine Learning: Progress and Futurescited by 40
  3. 2021A TinyML Platform for On-Device Continual Learning with Quantized Latent Replayscited by 37
  4. 2022Machine Learning for Microcontroller-Class Hardware: A Reviewcited by 28
  5. 2021Vega: A 10-Core SoC for IoT End-Nodes with DNN Acceleration and Cognitive Wake-Up From MRAM-Based State-Retentive Sleep Modecited by 16
  6. 2023Lightweight Neural Architecture Search for Temporal Convolutional Networks at the Edgecited by 15
  7. 2021ECG-TCN: Wearable Cardiac Arrhythmia Detection with a Temporal Convolutional Networkcited by 14
  8. 2022Q-PPG: Energy-Efficient PPG-based Heart Rate Monitoring on Wearable Devicescited by 14
  9. 2024On-Device Domain Learning for Keyword Spotting on Low-Power Extreme Edge Embedded Systemscited by 13
  10. 2023Flexible and Fully Quantized Ultra-Lightweight TinyissimoYOLO for Ultra-Low-Power Edge Systemscited by 11
  11. 2024Deeploy: Enabling Energy-Efficient Deployment of Small Language Models On Heterogeneous Microcontrollerscited by 11
  12. 2023Reduced Precision Floating-Point Optimization for Deep Neural Network On-Device Learning on MicroControllerscited by 10
38 more
  1. 2022Channel-wise Mixed-precision Assignment for DNN Inference on Constrained Edge Nodescited by 9
  2. 2022Distributed On-Sensor Compute System for AR/VR Devices: A Semi-Analytical Simulation Framework for Power Estimationcited by 8
  3. 2023Marsellus: A Heterogeneous RISC-V AI-IoT End-Node SoC with 2-to-8b DNN Acceleration and 30%-Boost Adaptive Body Biasingcited by 8
  4. 2024Optimizing the Deployment of Tiny Transformers on Low-Power MCUscited by 8
  5. 2023Enhancing Neural Architecture Search with Multiple Hardware Constraints for Deep Learning Model Deployment on Tiny IoT Devicescited by 7
  6. 2024A Tiny Transformer for Low-Power Arrhythmia Classification on Microcontrollerscited by 6
  7. 2024MATCH: Model-Aware TVM-based Compilation for Heterogeneous Edge Devicescited by 5
  8. 2022TCN Mapping Optimization for Ultra-Low Power Time-Series Edge Inferencecited by 4
  9. 2024Adaptive Deep Learning for Efficient Visual Pose Estimation aboard Ultra-low-power Nano-dronescited by 4
  10. 2024xTern: Energy-Efficient Ternary Neural Network Inference on RISC-V-Based Edge Systemscited by 4
  11. 2022HW-Aware Initialization of DNN Auto-Tuning to Improve Exploration Time and Robustnesscited by 3
  12. 2023Deep Neural Network Architecture Search for Accurate Visual Pose Estimation aboard Nano-UAVscited by 3
  13. 2023Precision-aware Latency and Energy Balancing on Multi-Accelerator Platforms for DNN Inferencecited by 3
  14. 2023Dynamic Decision Tree Ensembles for Energy-Efficient Inference on IoT Edge Nodescited by 3
  15. 2023Free Bits: Latency Optimization of Mixed-Precision Quantized Neural Networks on the Edgecited by 3
  16. 202412 mJ per Class On-Device Online Few-Shot Class-Incremental Learningcited by 3
  17. 2025A Map-free Deep Learning-based Framework for Gate-to-Gate Monocular Visual Navigation aboard Miniaturized Aerial Vehiclescited by 3
  18. 2025ELASTIC: Efficient Once For All Iterative Search for Object Detection on Microcontrollerscited by 3
  19. 2025TruncQuant: Truncation-Ready Quantization for DNNs with Flexible Weight Bit Precisioncited by 3
  20. 2022A Heterogeneous In-Memory Computing Cluster For Flexible End-to-End Inference of Real-World Deep Neural Networkscited by 2
  21. 2023Fused Depthwise Tiling for Memory Optimization in TinyML Deep Neural Network Inferencecited by 2
  22. 2023MEMA Runtime Framework: Minimizing External Memory Accesses for TinyML on Microcontrollerscited by 2
  23. 2024A Heterogeneous RISC-V based SoC for Secure Nano-UAV Navigationcited by 2
  24. 2024High-throughput Visual Nano-drone to Nano-drone Relative Localization using Onboard Fully Convolutional Networkscited by 2
  25. 2024MCU-MixQ: A HW/SW Co-optimized Mixed-precision Neural Network Design Framework for MCUscited by 2
  26. 2024Optimization and Deployment of Deep Neural Networks for PPG-based Blood Pressure Estimation Targeting Low-power Wearablescited by 2
  27. 2025EdgeMark: An Automation and Benchmarking System for Embedded Artificial Intelligence Toolscited by 2
  28. 2025MSQ: Memory-Efficient Bit Sparsification Quantizationcited by 2
  29. 2025MiCo: End-to-End Mixed Precision Neural Network Co-Exploration Framework for Edge AIcited by 2
  30. 2025An Open-Source HW-SW Co-Development Framework Enabling Efficient Multi-Accelerator Systemscited by 2
  31. 2021Joint Program and Layout Transformations to enable Convolutional Operators on Specialized Hardware based on Constraint Programmingcited by 1
  32. 2023Ultra-low Power Deep Learning-based Monocular Relative Localization Onboard Nano-quadrotorscited by 1
  33. 2023EpiDeNet: An Energy-Efficient Approach to Seizure Detection for Embedded Systemscited by 1
  34. 2024Optimized Deployment of Deep Neural Networks for Visual Pose Estimation on Nano-dronescited by 1
  35. 2024OpTC -- A Toolchain for Deployment of Neural Networks on AURIX TC3xx Microcontrollerscited by 1
  36. 2023A Survey on Design Methodologies for Accelerating Deep Learning on Heterogeneous Architecturescited by 0
  37. 2024Accelerating Depthwise Separable Convolutions on Ultra-Low-Power Devicescited by 0
  38. 2025MaRVIn: A Cross-Layer Mixed-Precision RISC-V Framework for DNN Inference, from ISA Extension to Hardware Accelerationcited by 0

50 most cited of 57 All 57 (JSON)