mlsanityData to 7 Oct 2026

An Empirical Study of Low Precision Quantization for TinyML

arXiv 2203.05492 ↗First posted 10 Mar 2022cs.LGcited by 6

Not on the map. Why?

Abstract (arXiv)

Tiny machine learning (tinyML) has emerged during the past few years aiming to deploy machine learning models to embedded AI processors with highly constrained memory and computation capacity. Low precision quantization is an important model compression technique that can greatly reduce both memory consumption and computation cost of model inference. In this study, we focus on post-training quantization (PTQ) algorithms that quantize a model to low-bit (less than 8-bit) precision with only a small set of calibration data and benchmark them on different tinyML use cases. To achieve a fair comparison, we build a simulated quantization framework to investigate recent PTQ algorithms. Furthermore, we break down those algorithms into essential components and re-assembled a generic PTQ pipeline. With ablation study on different alternatives of components in the pipeline, we reveal key design choices when performing low precision quantization. We hope this work could provide useful data points and shed lights on the future research of low precision quantization.

Key facts. Cited by 6 papers in the map (data through 7 Oct 2026). Builds on 22 of its 27 references that are in the map. First posted on arXiv 10 Mar 2022.

Builds on

22 of 27 references.

References · 22

  1. 2015Distilling the Knowledge in a Neural Networkcited by 10,076
  2. 2015Deep Compression: Compressing Deep Neural Networks with Pruning, Trained Quantization and Huffman Codingcited by 2,291
  3. 2013Estimating or Propagating Gradients Through Stochastic Neurons for Conditional Computationcited by 2,031
  4. 2018DARTS: Differentiable Architecture Searchcited by 1,782
  5. 2021A Survey of Quantization Methods for Efficient Neural Network Inferencecited by 798
  6. 2019Once-for-All: Train One Network and Specialize it for Efficient Deploymentcited by 778
  7. 2019Learned Step Size Quantizationcited by 604
  8. 2020Up or Down? Adaptive Rounding for Post-Training Quantizationcited by 600
  9. 2018Quantizing deep convolutional networks for efficient inference: A whitepapercited by 507
  10. 2021BRECQ: Pushing the Limit of Post-Training Quantization by Block Reconstructioncited by 464
  11. 2021A White Paper on Neural Network Quantizationcited by 453
  12. 2019Data-Free Quantization Through Weight Equalization and Bias Correctioncited by 328
10 more
  1. 2020MCUNet: Tiny Deep Learning on IoT Devicescited by 269
  2. 2019Differentiable Soft Quantization: Bridging Full-Precision and Low-Bit Neural Networkscited by 234
  3. 2020Integer Quantization for Deep Learning Inference: Principles and Empirical Evaluationcited by 183
  4. 2019Improving Neural Network Quantization without Retraining using Outlier Channel Splittingcited by 150
  5. 2021MLPerf Tiny Benchmarkcited by 118
  6. 2020Improving Post Training Neural Quantization: Layer-wise Calibration and Integer Programmingcited by 89
  7. 2019Trained Quantization Thresholds for Accurate and Efficient Fixed-Point Inference of Deep Neural Networkscited by 87
  8. 2021MCUNetV2: Memory-Efficient Patch-based Inference for Tiny Deep Learningcited by 87
  9. 2021Quantization and Deployment of Deep Neural Networks on Microcontrollerscited by 20
  10. 2021Quantization-Guided Training for Compact TinyML Modelscited by 10

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2023: 4; 2024: 0; 2025: 0; 2026: 2.

6 papers.

  1. 2023Intriguing Properties of Quantization at Scalecited by 48
  2. 2023A Machine Learning-oriented Survey on Tiny Machine Learningcited by 26
  3. 2023TinyReptile: TinyML with Federated Meta-Learningcited by 4
  4. 2026Quantizing Recursive Reasoning Modelscited by 2
  5. 2023Towards Clip-Free Quantized Super-Resolution Networks: How to Tame Representative Imagescited by 0
  6. 2026Late Breaking Results: Conversion of Neural Networks into Logic Flows for Edge Computingcited by 0