Weight Equalizing Shift Scaler-Coupled Post-training Quantization

• URL: http://arxiv.org/abs/2008.05767v1
• Date: Thu, 13 Aug 2020 09:19:57 GMT
• Title: Weight Equalizing Shift Scaler-Coupled Post-training Quantization
• Authors: Jihun Oh, SangJeong Lee, Meejeong Park, Pooni Walagaurav and Kiseok Kwon
• Abstract summary: Post-training, layer-wise quantization is preferable because it is free from retraining and is hardware-friendly. accuracy degradation has occurred when a neural network model has a big difference of per-out-channel weight ranges. We propose a new weight equalizing shift scaler, i.e. rescaling the weight range per channel by a 4-bit binary shift, prior to a layer-wise quantization.
• Score: 0.5936318628878774
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Post-training, layer-wise quantization is preferable because it is free from retraining and is hardware-friendly. Nevertheless, accuracy degradation has occurred when a neural network model has a big difference of per-out-channel weight ranges. In particular, the MobileNet family has a tragedy drop in top-1 accuracy from 70.60% ~ 71.87% to 0.1% on the ImageNet dataset after 8-bit weight quantization. To mitigate this significant accuracy reduction, we propose a new weight equalizing shift scaler, i.e. rescaling the weight range per channel by a 4-bit binary shift, prior to a layer-wise quantization. To recover the original output range, inverse binary shifting is efficiently fused to the existing per-layer scale compounding in the fixed-computing convolutional operator of the custom neural processing unit. The binary shift is a key feature of our algorithm, which significantly improved the accuracy performance without impeding the memory footprint. As a result, our proposed method achieved a top-1 accuracy of 69.78% ~ 70.96% in MobileNets and showed robust performance in varying network models and tasks, which is competitive to channel-wise quantization results.

Related papers

• Post-Training Quantization for Re-parameterization via Coarse & Fine Weight Splitting [13.270381125055275]
  We propose a coarse & fine weight splitting (CFWS) method to reduce quantization error of weight. We develop an improved KL metric to determine optimal quantization scales for activation. For example, the quantized RepVGG-A1 model exhibits a mere 0.3% accuracy loss.
  arXiv  Detail & Related papers  (2023-12-17T02:31:20Z)
• Quantized Neural Networks for Low-Precision Accumulation with Guaranteed Overflow Avoidance [68.8204255655161]
  We introduce a quantization-aware training algorithm that guarantees avoiding numerical overflow when reducing the precision of accumulators during inference. We evaluate our algorithm across multiple quantized models that we train for different tasks, showing that our approach can reduce the precision of accumulators while maintaining model accuracy with respect to a floating-point baseline.
  arXiv  Detail & Related papers  (2023-01-31T02:46:57Z)
• BiTAT: Neural Network Binarization with Task-dependent Aggregated Transformation [116.26521375592759]
  Quantization aims to transform high-precision weights and activations of a given neural network into low-precision weights/activations for reduced memory usage and computation. Extreme quantization (1-bit weight/1-bit activations) of compactly-designed backbone architectures results in severe performance degeneration. This paper proposes a novel Quantization-Aware Training (QAT) method that can effectively alleviate performance degeneration.
  arXiv  Detail & Related papers  (2022-07-04T13:25:49Z)
• Post-training Quantization for Neural Networks with Provable Guarantees [9.58246628652846]
  We modify a post-training neural-network quantization method, GPFQ, that is based on a greedy path-following mechanism. We prove that for quantizing a single-layer network, the relative square error essentially decays linearly in the number of weights.
  arXiv  Detail & Related papers  (2022-01-26T18:47:38Z)
• OMPQ: Orthogonal Mixed Precision Quantization [64.59700856607017]
  Mixed precision quantization takes advantage of hardware's multiple bit-width arithmetic operations to unleash the full potential of network quantization. We propose to optimize a proxy metric, the concept of networkity, which is highly correlated with the loss of the integer programming. This approach reduces the search time and required data amount by orders of magnitude, with little compromise on quantization accuracy.
  arXiv  Detail & Related papers  (2021-09-16T10:59:33Z)
• Post-training deep neural network pruning via layer-wise calibration [70.65691136625514]
  We propose a data-free extension of the approach for computer vision models based on automatically-generated synthetic fractal images. When using real data, we are able to get a ResNet50 model on ImageNet with 65% sparsity rate in 8-bit precision in a post-training setting.
  arXiv  Detail & Related papers  (2021-04-30T14:20:51Z)
• Learnable Companding Quantization for Accurate Low-bit Neural Networks [3.655021726150368]
  Quantizing deep neural networks is an effective method for reducing memory consumption and improving inference speed. It is still hard for extremely low-bit models to achieve accuracy comparable with that of full-precision models. We propose learnable companding quantization (LCQ) as a novel non-uniform quantization method for 2-, 3-, and 4-bit models.
  arXiv  Detail & Related papers  (2021-03-12T09:06:52Z)
• VS-Quant: Per-vector Scaled Quantization for Accurate Low-Precision Neural Network Inference [7.886868529510128]
  Quantization maps floating-point weights and activations in a trained model to low-bitwidth integer values using scale factors. Excessive quantization, reducing precision too aggressively, results in accuracy degradation. Per-vector scale factors can be implemented with low-bitwidth integers when using a two-level quantization scheme.
  arXiv  Detail & Related papers  (2021-02-08T19:56:04Z)
• Direct Quantization for Training Highly Accurate Low Bit-width Deep Neural Networks [73.29587731448345]
  This paper proposes two novel techniques to train deep convolutional neural networks with low bit-width weights and activations. First, to obtain low bit-width weights, most existing methods obtain the quantized weights by performing quantization on the full-precision network weights. Second, to obtain low bit-width activations, existing works consider all channels equally.
  arXiv  Detail & Related papers  (2020-12-26T15:21:18Z)
• Subtensor Quantization for Mobilenets [5.735035463793008]
  Quantization for deep neural networks (DNN) have enabled developers to deploy models with less memory and more efficient low-power inference. In this paper, we analyzed several root causes of quantization loss and proposed alternatives that do not rely on per-channel or training-aware approaches. We evaluate the image classification task on ImageNet dataset, and our post-training quantized 8-bit inference top-1 accuracy in within 0.7% of the floating point version.
  arXiv  Detail & Related papers  (2020-11-04T15:41:47Z)
• Widening and Squeezing: Towards Accurate and Efficient QNNs [125.172220129257]
  Quantization neural networks (QNNs) are very attractive to the industry because their extremely cheap calculation and storage overhead, but their performance is still worse than that of networks with full-precision parameters. Most of existing methods aim to enhance performance of QNNs especially binary neural networks by exploiting more effective training techniques. We address this problem by projecting features in original full-precision networks to high-dimensional quantization features.
  arXiv  Detail & Related papers  (2020-02-03T04:11:13Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.