FxP-QNet: A Post-Training Quantizer for the Design of Mixed Low-Precision DNNs with Dynamic Fixed-Point Representation
Not on the map. Why?
Abstract (arXiv)
Deep neural networks (DNNs) have demonstrated their effectiveness in a wide range of computer vision tasks, with the state-of-the-art results obtained through complex and deep structures that require intensive computation and memory. Now-a-days, efficient model inference is crucial for consumer applications on resource-constrained platforms. As a result, there is much interest in the research and development of dedicated deep learning (DL) hardware to improve the throughput and energy efficiency of DNNs. Low-precision representation of DNN data-structures through quantization would bring great benefits to specialized DL hardware. However, the rigorous quantization leads to a severe accuracy drop. As such, quantization opens a large hyper-parameter space at bit-precision levels, the exploration of which is a major challenge. In this paper, we propose a novel framework referred to as the Fixed-Point Quantizer of deep neural Networks (FxP-QNet) that flexibly designs a mixed low-precision DNN for integer-arithmetic-only deployment. Specifically, the FxP-QNet gradually adapts the quantization level for each data-structure of each layer based on the trade-off between the network accuracy and the low-precision requirements. Additionally, it employs post-training self-distillation and network prediction error statistics to optimize the quantization of floating-point values into fixed-point numbers. Examining FxP-QNet on state-of-the-art architectures and the benchmark ImageNet dataset, we empirically demonstrate the effectiveness of FxP-QNet in achieving the accuracy-compression trade-off without the need for training. The results show that FxP-QNet-quantized AlexNet, VGG-16, and ResNet-18 reduce the overall memory requirements of their full-precision counterparts by 7.16x, 10.36x, and 6.44x with less than 0.95%, 0.95%, and 1.99% accuracy drop, respectively.
Key facts. Cited by 2 papers in the map (data through 7 Oct 2026). Builds on 63 of its 100 references that are in the map. First posted on arXiv 22 Mar 2022.
Builds on
63 of 100 references.
References · 63
- 2015Deep Residual Learning for Image Recognitioncited by 51,809
- 2014Very Deep Convolutional Networks for Large-Scale Image Recognitioncited by 15,267
- 2015Distilling the Knowledge in a Neural Networkcited by 10,076
- 2015Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networkscited by 9,617
- 2014ImageNet Large Scale Visual Recognition Challengecited by 8,441
- 2017Mask R-CNNcited by 7,259
- 2015Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shiftcited by 6,979
- 2016Densely Connected Convolutional Networkscited by 6,691
- 2014Fully Convolutional Networks for Semantic Segmentationcited by 4,845
- 2014Going Deeper with Convolutionscited by 4,231
- 2015Deep Compression: Compressing Deep Neural Networks with Pruning, Trained Quantization and Huffman Codingcited by 2,291
- 2013Rich feature hierarchies for accurate object detection and semantic segmentationcited by 2,160
38 more
- 2017Quantization and Training of Neural Networks for Efficient Integer-Arithmetic-Only Inferencecited by 1,548
- 2017Mixed Precision Trainingcited by 1,135
- 2016XNOR-Net: ImageNet Classification Using Binary Convolutional Neural Networkscited by 829
- 2019Learned Step Size Quantizationcited by 604
- 2016DoReFa-Net: Training Low Bitwidth Convolutional Neural Networks with Low Bitwidth Gradientscited by 557
- 2018Quantizing deep convolutional networks for efficient inference: A whitepapercited by 507
- 2018PACT: Parameterized Clipping Activation for Quantized Neural Networkscited by 456
- 2017Efficient Processing of Deep Neural Networks: A Tutorial and Surveycited by 427
- 2018HAQ: Hardware-Aware Automated Quantization with Mixed Precisioncited by 426
- 2016Quantized Neural Networks: Training Neural Networks with Low Precision Weights and Activationscited by 413
- 2019HAWQ: Hessian AWare Quantization of Neural Networks with Mixed-Precisioncited by 387
- 2015Deep Learning with Limited Numerical Precisioncited by 368
- 2016EIE: Efficient Inference Engine on Compressed Deep Neural Networkcited by 291
- 2014Caffe: Convolutional Architecture for Fast Feature Embeddingcited by 287
- 2018LQ-Nets: Learned Quantization for Highly Accurate and Compact Deep Neural Networkscited by 272
- 2017Incremental Network Quantization: Towards Lossless CNNs with Low-Precision Weightscited by 211
- 2016Trained Ternary Quantizationcited by 203
- 2016Ternary Weight Networkscited by 195
- 2019Low-bit Quantization of Neural Networks for Efficient Inferencecited by 185
- 2020Binary Neural Networks: A Surveycited by 184
- 2018Learning to Quantize Deep Networks by Optimizing Quantization Intervals with Task Losscited by 157
- 2018Mixed Precision Quantization of ConvNets via Differentiable Neural Architecture Searchcited by 122
- 2017Deep Learning with Low Precision by Half-wave Gaussian Quantizationcited by 115
- 2017Apprentice: Using Knowledge Distillation Techniques To Improve Low-Precision Network Accuracycited by 100
- 2016Convolutional Neural Networks using Logarithmic Data Representationcited by 97
- 2017Bit Fusion: Bit-Level Dynamically Composable Architecture for Accelerating Deep Neural Networkscited by 95
- 2019Are All Layers Created Equal?cited by 94
- 2017Towards Effective Low-bitwidth Convolutional Neural Networkscited by 76
- 2016Loss-aware Binarization of Deep Networkscited by 56
- 2016SqueezeDet: Unified, Small, Low Power Fully Convolutional Neural Networks for Real-Time Object Detection for Autonomous Drivingcited by 55
- 2020PROFIT: A Novel Training Method for sub-4-bit MobileNet Modelscited by 50
- 2017WRPN: Wide Reduced-Precision Networkscited by 49
- 2019Same, Same But Different - Recovering Neural Network Quantization Error Through Weight Factorizationcited by 45
- 2018Post-training 4-bit quantization of convolution networks for rapid-deploymentcited by 38
- 2015Resiliency of Deep Neural Networks under Quantizationcited by 34
- 2018Loss-aware Weight Quantization of Deep Networkscited by 31
- 2019Memory-Driven Mixed Low Precision Quantization For Enabling Deep Network Inference On Microcontrollerscited by 31
- 2019FPGA-based Accelerators of Deep Learning Networks for Learning and Classification: A Reviewcited by 29
50 most cited of 63 All 63 (JSON)
Built on by
Citing papers per year: 2022: 1; 2023: 1.
2 papers.