Saltar a contenido

Bibliografía

Revisada en noviembre de 2026. [NV] indica que no se pudo verificar algún dato (venue o existencia); compruébalo antes de citar.

Surveys y revisiones

  • Lin, J., Zhu, L., Chen, W.-M., Wang, W.-C. y Han, S. (2023). "Tiny Machine Learning: Progress and Futures". IEEE Circuits and Systems Magazine, 23(3), 8-34. DOI · arXiv. Empieza por aquí.
  • Heydari y Mahmoud (2025). "Tiny Machine Learning and On-Device Inference: A Survey". Sensors. Acceso abierto.
  • "From Tiny Machine Learning to Tiny Deep Learning: A Survey" (2025). arXiv 2506.18927. Venue [NV].
  • "Quantized Neural Networks for Microcontrollers: A Comprehensive Review of Methods, Platforms, and Applications" (2025). arXiv 2508.15008.
  • Yan, Koch y Sinnen (2024). "A survey on FPGA-based accelerator for ML models". arXiv 2412.15666. Venue [NV].
  • Gholami et al. (2021). "A Survey of Quantization Methods for Efficient Neural Network Inference". arXiv 2103.13630.
  • Pavan et al. (2026). Survey de aprendizaje en el dispositivo ("What changes after deployment?"). arXiv 2605.31226.

Fundamentos de compresión

  • Jacob, B. et al. (2018). "Quantization and Training of Neural Networks for Efficient Integer-Arithmetic-Only Inference". CVPR, 2704-2713. DOI · arXiv.
  • Han, S., Mao, H. y Dally, W. J. (2016). "Deep Compression: Compressing Deep Neural Networks with Pruning, Trained Quantization and Huffman Coding". ICLR. arXiv.

TinyML en microcontroladores

  • Lin, J. et al. (2020). "MCUNet: Tiny Deep Learning on IoT Devices". NeurIPS. arXiv.
  • Lin, J. et al. (2021). "MCUNetV2: Memory-Efficient Patch-based Inference for Tiny Deep Learning". NeurIPS. arXiv.
  • Lin, J. et al. (2022). "On-Device Training Under 256KB Memory". NeurIPS. arXiv.
  • David, R. et al. (2021). "TensorFlow Lite Micro: Embedded Machine Learning on TinyML Systems". MLSys [NV venue]. arXiv.
  • Banbury, C. et al. (2021). "MLPerf Tiny Benchmark". NeurIPS Datasets and Benchmarks [NV venue]. arXiv.
  • Lai, L., Suda, N. y Chandra, V. (2018). "CMSIS-NN: Efficient Neural Network Kernels for Arm Cortex-M CPUs". arXiv.

ML en FPGA: herramientas

  • Umuroglu, Y. et al. (2017). "FINN: A Framework for Fast, Scalable Binarized Neural Network Inference". FPGA'17. arXiv.
  • Blott, M. et al. (2018). "FINN-R: An End-to-End Deep-Learning Framework for Fast Exploration of Quantized Neural Networks". ACM TRETS, 11(3). DOI · arXiv.
  • Duarte, J. et al. (2018). "Fast inference of deep neural networks in FPGAs for particle physics". JINST, 13, P07027. DOI · arXiv.
  • Schulte, J.-F. et al. (2025). "hls4ml: A Flexible, Open-Source Platform for Deep Learning Acceleration on Reconfigurable Hardware". arXiv 2512.01463.
  • Pappalardo, A. et al. (2022). "QONNX: Representing Arbitrary-Precision Quantized Neural Networks". arXiv.
  • Borras, H. et al. (2022). "Open-source FPGA-ML codesign for the MLPerf Tiny Benchmark". arXiv. Usa PYNQ-Z2 (Zynq-7020).

Redes basadas en LUT y lógica

  • Umuroglu, Y., Akhauri, Y., Fraser, N. J. y Blott, M. (2020). "LogicNets: Co-Designed Neural Networks and Circuits for Extreme-Throughput Applications". FPL, 291-297. arXiv.
  • Andronic, M. y Constantinides, G. A. (2023). "PolyLUT: Learning Piecewise Polynomials for Ultra-Low Latency FPGA LUT-based Inference". ICFPT, 60-68. arXiv. Versión extendida en IEEE Transactions on Computers, 74(9), 2025.
  • Andronic, M. y Constantinides, G. A. (2024). "NeuraLUT: Hiding Neural Network Density in Boolean Synthesizable Functions". FPL, 140-148. arXiv.
  • Andronic y Constantinides (2025). NeuraLUT-Assemble. arXiv 2504.00592. Venue [NV].
  • PolyLUT con poda estructurada (2025). arXiv 2501.08043.
  • Petersen, F. et al. (2022). "Deep Differentiable Logic Gate Networks". NeurIPS. arXiv.
  • Petersen, F. et al. (2024). "Convolutional Differentiable Logic Gate Networks". NeurIPS (oral). arXiv.
  • ReducedLUT (2024). arXiv 2412.18579.
  • Preprints 2025-2026: WARP-LUTs (arXiv 2510.15655), BitLogic (arXiv 2602.07400), FPGN (arXiv 2607.08427), recursos de redes de compuertas lógicas en FPGA (arXiv 2605.04109).

Selección de rasgos y conjuntos aproximados en sistemas embebidos

  • Nesterenko, M. y Lewis, R. (2022). "Rough Sets for Intelligence on Embedded Systems". Springer. DOI.
  • Kopczynski, M. (2026). "Scalable FPGA Hardware Architecture for Parallel Reduct Computation in Big Datasets Using Rough Sets". International Journal of Electronics and Telecommunications, 72(3). Enlace.
  • Artículo de 2025 que señala la selección de rasgos como poco estudiada en TinyML: MDPI AI 6(6):107.

Radar y TinyML

  • Strobel, M. et al. (2024). "Gesture Recognition for FMCW Radar on the Edge". arXiv.
  • Ronco, Schilk y Magno (2024). "TinyssimoRadar". ETH Research Collection.
  • Scherer, M. et al. (2020). "TinyRadarNN". arXiv.
  • Survey de nowcasting de precipitación con aprendizaje profundo: arXiv 2406.04867.