mlsanityData to 7 Oct 2026

Elastic Significant Bit Quantization and Acceleration for Deep Neural Networks

arXiv 2109.03513 ↗First posted 8 Sep 2021cs.CV eess.SPcited by 4

Not on the map. Why?

Abstract (arXiv)

Quantization has been proven to be a vital method for improving the inference efficiency of deep neural networks (DNNs). However, it is still challenging to strike a good balance between accuracy and efficiency while quantizing DNN weights or activation values from high-precision formats to their quantized counterparts. We propose a new method called elastic significant bit quantization (ESB) that controls the number of significant bits of quantized values to obtain better inference accuracy with fewer resources. We design a unified mathematical formula to constrain the quantized values of the ESB with a flexible number of significant bits. We also introduce a distribution difference aligner (DDA) to quantitatively align the distributions between the full-precision weight or activation values and quantized values. Consequently, ESB is suitable for various bell-shaped distributions of weights and activation of DNNs, thus maintaining a high inference accuracy. Benefitting from fewer significant bits of quantized values, ESB can reduce the multiplication complexity. We implement ESB as an accelerator and quantitatively evaluate its efficiency on FPGAs. Extensive experimental results illustrate that ESB quantization consistently outperforms state-of-the-art methods and achieves average accuracy improvements of 4.78%, 1.92%, and 3.56% over AlexNet, ResNet18, and MobileNetV2, respectively. Furthermore, ESB as an accelerator can achieve 10.95 GOPS peak performance of 1k LUTs without DSPs on the Xilinx ZCU102 FPGA platform. Compared with CPU, GPU, and state-of-the-art accelerators on FPGAs, the ESB accelerator can improve the energy efficiency by up to 65x, 11x, and 26x, respectively.

Key facts. Cited by 4 papers in the map (data through 7 Oct 2026). Builds on 29 of its 48 references that are in the map. First posted on arXiv 8 Sep 2021.

Builds on

29 of 48 references.

References · 29

  1. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  2. 2015Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shiftcited by 6,979
  3. 2018MobileNetV2: Inverted Residuals and Linear Bottleneckscited by 4,773
  4. 2015Deep Compression: Compressing Deep Neural Networks with Pruning, Trained Quantization and Huffman Codingcited by 2,291
  5. 2017Quantization and Training of Neural Networks for Efficient Integer-Arithmetic-Only Inferencecited by 1,548
  6. 2016XNOR-Net: ImageNet Classification Using Binary Convolutional Neural Networkscited by 829
  7. 2016DoReFa-Net: Training Low Bitwidth Convolutional Neural Networks with Low Bitwidth Gradientscited by 557
  8. 2018PACT: Parameterized Clipping Activation for Quantized Neural Networkscited by 456
  9. 2018HAQ: Hardware-Aware Automated Quantization with Mixed Precisioncited by 426
  10. 2014Deeply-Supervised Netscited by 364
  11. 2018LQ-Nets: Learned Quantization for Highly Accurate and Compact Deep Neural Networkscited by 272
  12. 2019Differentiable Soft Quantization: Bridging Full-Precision and Low-Bit Neural Networkscited by 234
17 more
  1. 2017Incremental Network Quantization: Towards Lossless CNNs with Low-Precision Weightscited by 211
  2. 2018Learning to Quantize Deep Networks by Optimizing Quantization Intervals with Task Losscited by 157
  3. 2015Compressing Neural Networks with the Hashing Trickcited by 150
  4. 2017Towards Accurate Binary Convolutional Neural Networkcited by 148
  5. 2019Additive Powers-of-Two Quantization: An Efficient Non-uniform Discretization for Neural Networkscited by 143
  6. 2016Convolutional Neural Networks using Logarithmic Data Representationcited by 97
  7. 2020PatDNN: Achieving Real-Time DNN Execution on Mobile Devices with Pattern-based Weight Pruningcited by 75
  8. 2017Extremely Low Bit Neural Network: Squeeze the Last Bit Out with ADMMcited by 60
  9. 2019AutoQ: Automated Kernel-Wise Neural Network Quantizationcited by 60
  10. 2017Flexpoint: An Adaptive Numerical Format for Efficient Training of Deep Neural Networkscited by 43
  11. 2018SYQ: Learning Symmetric Quantization For Efficient Deep Neural Networkscited by 24
  12. 2018Blended Coarse Gradient Descent for Full Quantization of Deep Neural Networkscited by 20
  13. 2020VecQ: Minimal Loss DNN Model Compression With Vectorized Weight Quantizationcited by 15
  14. 2018Deep Positron: A Deep Neural Network Using the Posit Number Systemcited by 13
  15. 2018Design Flow of Accelerating Hybrid Extremely Low Bit-width Neural Network in Embedded FPGAcited by 6
  16. 2019AdaptivFloat: A Floating-point based Data Type for Resilient Deep Learning Inferencecited by 5
  17. 2019AddNet: Deep Neural Networks Using FPGA-Optimized Multiplierscited by 5

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2024: 3; 2025: 0; 2026: 1.

4 papers.

  1. 2024Deep Feature Surgery: Towards Accurate and Efficient Multi-Exit Networkscited by 2
  2. 2024Minimize Quantization Output Error with Bias Compensationcited by 1
  3. 2024Accelerating PoT Quantization on Edge Devicescited by 1
  4. 2026PoTAcc: A Pipeline for End-to-End Acceleration of Power-of-Two Quantized DNNscited by 0