Propagating Asymptotic-Estimated Gradients for Low Bitwidth Quantized Neural Networks

• URL: http://arxiv.org/abs/2003.04296v1
• Date: Wed, 4 Mar 2020 03:17:47 GMT
• Title: Propagating Asymptotic-Estimated Gradients for Low Bitwidth Quantized Neural Networks
• Authors: Jun Chen, Yong Liu, Hao Zhang, Shengnan Hou, Jian Yang
• Abstract summary: We propose a novel Asymptotic-Quantized Estimator (AQE) to estimate the gradient. At the end of training, the weights and activations have been quantized to low-precision. In the inference phase, we can use XNOR or SHIFT operations instead of convolution operations to accelerate the MINW-Net.
• Score: 31.168156284218746
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: The quantized neural networks (QNNs) can be useful for neural network acceleration and compression, but during the training process they pose a challenge: how to propagate the gradient of loss function through the graph flow with a derivative of 0 almost everywhere. In response to this non-differentiable situation, we propose a novel Asymptotic-Quantized Estimator (AQE) to estimate the gradient. In particular, during back-propagation, the graph that relates inputs to output remains smoothness and differentiability. At the end of training, the weights and activations have been quantized to low-precision because of the asymptotic behaviour of AQE. Meanwhile, we propose a M-bit Inputs and N-bit Weights Network (MINW-Net) trained by AQE, a quantized neural network with 1-3 bits weights and activations. In the inference phase, we can use XNOR or SHIFT operations instead of convolution operations to accelerate the MINW-Net. Our experiments on CIFAR datasets demonstrate that our AQE is well defined, and the QNNs with AQE perform better than that with Straight-Through Estimator (STE). For example, in the case of the same ConvNet that has 1-bit weights and activations, our MINW-Net with AQE can achieve a prediction accuracy 1.5\% higher than the Binarized Neural Network (BNN) with STE. The MINW-Net, which is trained from scratch by AQE, can achieve comparable classification accuracy as 32-bit counterparts on CIFAR test sets. Extensive experimental results on ImageNet dataset show great superiority of the proposed AQE and our MINW-Net achieves comparable results with other state-of-the-art QNNs.

Related papers

• Histogram-Equalized Quantization for logic-gated Residual Neural Networks [2.7036595757881323]
  Histogram-Equalized Quantization (HEQ) is an adaptive framework for linear symmetric quantization. HEQ automatically adapts the quantization thresholds using a unique step size optimization. Experiments on the STL-10 dataset even show that HEQ enables a proper training of our proposed logic-gated (OR, MUX) residual networks.
  arXiv  Detail & Related papers  (2025-01-08T14:06:07Z)
• Projected Stochastic Gradient Descent with Quantum Annealed Binary Gradients [51.82488018573326]
  We present QP-SBGD, a novel layer-wise optimiser tailored towards training neural networks with binary weights. BNNs reduce the computational requirements and energy consumption of deep learning models with minimal loss in accuracy. Our algorithm is implemented layer-wise, making it suitable to train larger networks on resource-limited quantum hardware.
  arXiv  Detail & Related papers  (2023-10-23T17:32:38Z)
• Designing strong baselines for ternary neural network quantization through support and mass equalization [7.971065005161565]
  Deep neural networks (DNNs) offer the highest performance in a wide range of applications in computer vision. This computational burden can be dramatically reduced by quantizing floating point values to ternary values. We show experimentally that our approach allows to significantly improve the performance of ternary quantization through a variety of scenarios.
  arXiv  Detail & Related papers  (2023-06-30T07:35:07Z)
• Neural Capacitance: A New Perspective of Neural Network Selection via Edge Dynamics [85.31710759801705]
  Current practice requires expensive computational costs in model training for performance prediction. We propose a novel framework for neural network selection by analyzing the governing dynamics over synaptic connections (edges) during training. Our framework is built on the fact that back-propagation during neural network training is equivalent to the dynamical evolution of synaptic connections.
  arXiv  Detail & Related papers  (2022-01-11T20:53:15Z)
• A White Paper on Neural Network Quantization [20.542729144379223]
  We introduce state-of-the-art algorithms for mitigating the impact of quantization noise on the network's performance. We consider two main classes of algorithms: Post-Training Quantization (PTQ) and Quantization-Aware-Training (QAT)
  arXiv  Detail & Related papers  (2021-06-15T17:12:42Z)
• Toward Trainability of Quantum Neural Networks [87.04438831673063]
  Quantum Neural Networks (QNNs) have been proposed as generalizations of classical neural networks to achieve the quantum speed-up. Serious bottlenecks exist for training QNNs due to the vanishing with gradient rate exponential to the input qubit number. We show that QNNs with tree tensor and step controlled structures for the application of binary classification. Simulations show faster convergent rates and better accuracy compared to QNNs with random structures.
  arXiv  Detail & Related papers  (2020-11-12T08:32:04Z)
• A Statistical Framework for Low-bitwidth Training of Deep Neural Networks [70.77754244060384]
  Fully quantized training (FQT) uses low-bitwidth hardware by quantizing the activations, weights, and gradients of a neural network model. One major challenge with FQT is the lack of theoretical understanding, in particular of how gradient quantization impacts convergence properties.
  arXiv  Detail & Related papers  (2020-10-27T13:57:33Z)
• AQD: Towards Accurate Fully-Quantized Object Detection [94.06347866374927]
  We propose an Accurate Quantized object Detection solution, termed AQD, to get rid of floating-point computation. Our AQD achieves comparable or even better performance compared with the full-precision counterpart under extremely low-bit schemes.
  arXiv  Detail & Related papers  (2020-07-14T09:07:29Z)
• A Learning Framework for n-bit Quantized Neural Networks toward FPGAs [20.83904734716565]
  This paper proposes a novel learning framework for n-bit QNNs, whose weights are constrained to the power of two. We also propose a novel QNN structure named n-BQ-NN, which uses shift operation to replace the multiply operation. Experiments show that our n-BQ-NN with our SVPE can execute 2.9 times faster than with the vector processing element (VPE) in inference.
  arXiv  Detail & Related papers  (2020-04-06T04:21:24Z)
• 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)
• Training of Quantized Deep Neural Networks using a Magnetic Tunnel Junction-Based Synapse [23.08163992580639]
  Quantized neural networks (QNNs) are being actively researched as a solution for the computational complexity and memory intensity of deep neural networks. We show how magnetic tunnel junction (MTJ) devices can be used to support QNN training. We introduce a novel synapse circuit that uses the MTJ behavior to support the quantize update.
  arXiv  Detail & Related papers  (2019-12-29T11:36:32Z)

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.