Researches have demonstrated that low bit-width (e.g., INT8) quantization can be employed to accelerate the inference process. It makes the gradient quantization very promising since the backward propagation requires approximately twice more computation than forward one. Due to the variability and uncertainty of gradient distribution, a lot of methods have been proposed to attain training stability. However, most of them ignore the channel-wise gradient distributions and the impact of gradients with different magnitudes, resulting in the degradation of final accuracy. In this paper, we propose a novel INT8 quantization training framework for convolutional neural network to address the above issues. Specifically, we adopt Gradient Vectorized Quantization to quantize the gradient, based on the observation that layer-wise gradients contain multiple distributions along the channel dimension. Then, Magnitude-aware Clipping Strategy is introduced by taking the magnitudes of gradients into consideration when minimizing the quantization error, and we present a theoretical derivation to solve the quantization parameters of different distributions. Experimental results on broad range of computer vision tasks, such as image classification, object detection and video classification, demonstrate that the proposed Distribution Adaptive INT8 Quantization training method has achieved almost lossless training accuracy for different backbones, including ResNet, MobileNetV2, InceptionV3, VGG and AlexNet, which is superior to the state-of-the-art techniques. Moreover, we further implement the INT8 kernel that can accelerate the training iteration more than 200% under the latest Turing architecture, i.e., our method excels on both training accuracy and speed.
Deep convolutional neural networks (CNN) has become the most promising method for object recognition, repeatedly demonstrating record breaking results for image classification and object detection in recent years. However, a very deep CNN generally involves many layers with millions of parameters, making the storage of the network model to be extremely large. This prohibits the usage of deep CNNs on resource limited hardware, especially cell phones or other embedded devices. In this paper, we tackle this model storage issue by investigating information theoretical vector quantization methods f…
Graph Convolutional Networks (GCNs) are widely used in a variety of applications, and can be seen as an unstructured version of standard Convolutional Neural Networks (CNNs). As in CNNs, the computational cost of GCNs for large input graphs (such as large point clouds or meshes) can be high and inhibit the use of these networks, especially in environments with low computational resources. To ease these costs, quantization can be applied to GCNs. However, aggressive quantization of the feature maps can lead to a significant degradation in performance. On a different note, Haar wavelet transform…
Beijing University of Chemical Technology（北京化工大学）；College of Information Science & Technology（信息科学与技术学院）
For the task of subdecimeter aerial imagery segmentation, fine-grained semantic segmentation results are usually difficult to obtain because of complex remote sensing content and optical conditions. Recently, convolutional neural networks (CNNs) have shown outstanding performance on this task. Although many deep neural network structures and techniques have been applied to improve the accuracy, few have paid attention to better differentiating the easily confused classes. In this paper, we propose TreeSegNet which adopts an adaptive network to increase the classification rate at the pixelwise …
Tsinghua University（清华大学）；The Chinese University of Hong Kong（香港中文大学）；SenseTime（商汤科技）；The University of Sydney（悉尼大学）
In this paper, we propose a simple and general framework for training very tiny CNNs for object detection. Due to limited representation ability, it is challenging to train very tiny networks for complicated tasks like detection. To the best of our knowledge, our method, called Quantization Mimic, is the first one focusing on very tiny networks. We utilize two types of acceleration methods: mimic and quantization. Mimic improves the performance of a student network by transfering knowledge from a teacher network. Quantization converts a full-precision network to a quantized one without large d…
ACM-UNet: Adaptive Integration of CNNs and Mamba for Efficient Medical Image Segmentation
Wuhan University of Technology（武汉理工大学）；Chinese Academy of Medical Sciences and Peking Union Medical College（中国医学科学院北京协和医学院）；Zhongnan Hospital of Wuhan University（武汉大学中南医院）
Efficient 3D affinely equivariant CNNs with adaptive fusion of augmented spherical Fourier-Bessel bases
School of Computer and Artificial Intelligence, Nanjing University of Finance and Economics（南京金融学院计算机与人工智能学院）；Interdisciplinary Center for Scientific Computing, Mannheim Institute for Intelligent Systems in Medicine, Medical Faculty Mannheim, Heidelberg University（科学计算交叉中心、医学智能系统研究所、曼海姆医学院、海德堡大学）；Computer Assisted Clinical Medicine, Mannheim Institute for Intelligent Systems in Medicine, Medical Faculty Mannheim, Heidelberg University（临床医学辅助计算机、医学智能系统研究所、曼海姆医学院、海德堡大学）；Department of Diagnostic and Interventional Radiology, University Hospital Bonn（诊断与介入放射科、波恩大学医院）；Interdisciplinary Center for Scientific Computing, Central Institute for Computer Engineering, CSZ Heidelberg Center for Model-Based AI, Data Analysis and Modeling in Medicine, Mannheim Institute for Intelligent Systems in Medicine, Medical Faculty Mannheim, Heidelberg University（科学计算交叉中心、计算机工程中央研究所、海德堡模型驱动AI中心、医学数据与建模中心、曼海姆医学智能系统研究所、曼海姆医学院、海德堡大学）
AdaLog: Post-Training Quantization for Vision Transformers with Adaptive Logarithm Quantizer
State Key Laboratory of Virtual Reality Technology and Systems, Beihang University（北京航空航天大学虚拟现实技术与系统国家重点实验室）；School of Computer Science and Engineering, Beihang University（北京航空航天大学计算机科学与工程学院）
Accurate segmentation of the prostate from magnetic resonance (MR) images provides useful information for prostate cancer diagnosis and treatment. However, automated prostate segmentation from 3D MR images still faces several challenges. For instance, a lack of clear edge between the prostate and other anatomical structures makes it challenging to accurately extract the boundaries. The complex background texture and large variation in size, shape and intensity distribution of the prostate itself make segmentation even further complicated. With deep learning, especially convolutional neural net…
Recent work in network quantization produced state-of-the-art results using mixed precision quantization. An imperative requirement for many efficient edge device hardware implementations is that their quantizers are uniform and with power-of-two thresholds. In this work, we introduce the Hardware Friendly Mixed Precision Quantization Block (HMQ) in order to meet this requirement. The HMQ is a mixed precision quantization block that repurposes the Gumbel-Softmax estimator into a smooth estimator of a pair of quantization parameters, namely, bit-width and threshold. HMQs use this to search over…
Convolutional Neural Networks (CNNs) are known for requiring extensive computational resources, and quantization is among the best and most common methods for compressing them. While aggressive quantization (i.e., less than 4-bits) performs well for classification, it may cause severe performance degradation in image-to-image tasks such as semantic segmentation and depth estimation. In this paper, we propose Wavelet Compressed Convolution (WCC) -- a novel approach for high-resolution activation maps compression integrated with point-wise convolutions, which are the main computational cost of m…
Imperial College London（伦敦帝国学院）；Arm ML Research Lab（Arm机器学习研究实验室）
Modern convolutional neural networks (CNNs) are known to be overconfident in terms of their calibration on unseen input data. That is to say, they are more confident than they are accurate. This is undesirable if the probabilities predicted are to be used for downstream decision making. When considering accuracy, CNNs are also surprisingly robust to compression techniques, such as quantization, which aim to reduce computational and memory costs. We show that this robustness can be partially explained by the calibration behavior of modern CNNs, and may be improved with overconfidence. This is d…
Eindhoven University of Technology（埃因霍温理工大学）；Philips Innovation Services（飞利浦创新服务）
Convolutional Neural Networks (CNNs) have proven to be a powerful state-of-the-art method for image classification tasks. One drawback however is the high computational complexity and high memory consumption of CNNs which makes them unfeasible for execution on embedded platforms which are constrained on physical resources needed to support CNNs. Quantization has often been used to efficiently optimize CNNs for memory and computational complexity at the cost of a loss of prediction accuracy. We therefore propose a method to optimally quantize the weights, biases and activations of each layer of…
This paper discusses three basic blocks for the inference of convolutional neural networks (CNNs). Pyramid Vector Quantization (PVQ) is discussed as an effective quantizer for CNNs weights resulting in highly sparse and compressible networks. Properties of PVQ are exploited for the elimination of multipliers during inference while maintaining high performance. The result is then extended to any other quantized weights. The Tiny Yolo v3 CNN is used to compare such basic blocks.
This paper proposes a crowd counting method. Crowd counting is difficult because of large appearance changes of a target which caused by density and scale changes. Conventional crowd counting methods generally utilize one predictor (e,g., regression and multi-class classifier). However, such only one predictor can not count targets with large appearance changes well. In this paper, we propose to predict the number of targets using multiple CNNs specialized to a specific appearance, and those CNNs are adaptively selected according to the appearance of a test image. By integrating the selected C…
University of Coimbra（科英布拉大学）；Instituto de Telecomunicações（电信研究所）；ETH Zürich（苏黎世联邦理工学院）
In recent years, Convolutional Neural Networks (CNNs) have become the standard class of deep neural network for image processing, classification and segmentation tasks. However, the large strides in accuracy obtained by CNNs have been derived from increasing the complexity of network topologies, which incurs sizeable performance and energy penalties in the training and inference of CNNs. Many recent works have validated the effectiveness of parameter quantization, which consists in reducing the bit width of the network's parameters, to enable the attainment of considerable performance and ener…
Robert Bosch GmbH（罗伯特·博世有限公司）；University of Lübeck（吕贝克大学）
Unstructured pruning is well suited to reduce the memory footprint of convolutional neural networks (CNNs), both at training and inference time. CNNs contain parameters arranged in $K \times K$ filters. Standard unstructured pruning (SP) reduces the memory footprint of CNNs by setting filter elements to zero, thereby specifying a fixed subspace that constrains the filter. Especially if pruning is applied before or during training, this induces a strong bias. To overcome this, we introduce interspace pruning (IP), a general tool to improve existing pruning methods. It uses filters represented i…
This paper presents incremental network quantization (INQ), a novel method, targeting to efficiently convert any pre-trained full-precision convolutional neural network (CNN) model into a low-precision version whose weights are constrained to be either powers of two or zero. Unlike existing methods which are struggled in noticeable accuracy loss, our INQ has the potential to resolve this issue, as benefiting from two innovations. On one hand, we introduce three interdependent operations, namely weight partition, group-wise quantization and re-training. A well-proven measure is employed to divi…
University of Waterloo（滑铁卢大学）；Waterloo Artificial Intelligence Institute（滑铁卢人工智能研究所）
As the "Mobile AI" revolution continues to grow, so does the need to understand the behaviour of edge-deployed deep neural networks. In particular, MobileNets are the go-to family of deep convolutional neural networks (CNN) for mobile. However, they often have significant accuracy degradation under post-training quantization. While studies have introduced quantization-aware training and other methods to tackle this challenge, there is limited understanding into why MobileNets (and potentially depthwise-separable CNNs (DWSCNN) in general) quantize so poorly compared to other CNN architectures. …
INT8 quantization has become one of the standard techniques for deploying convolutional neural networks (CNNs) on edge devices to reduce the memory and computational resource usages. By analyzing quantized performances of existing mobile-target network architectures, we can raise an issue regarding the importance of network architecture for optimal INT8 quantization. In this paper, we present a new network architecture search (NAS) procedure to find a network that guarantees both full-precision (FLOAT32) and quantized (INT8) performances. We first propose critical but straightforward optimizat…
College of Biomedical Engineering & Instrument Science（生物医学工程与仪器科学学院）；Zhejiang University（浙江大学）
Data-free quantization is a task that compresses the neural network to low bit-width without access to original training data. Most existing data-free quantization methods cause severe performance degradation due to inaccurate activation clipping range and quantization error, especially for low bit-width. In this paper, we present a simple yet effective data-free quantization method with accurate activation clipping and adaptive batch normalization. Accurate activation clipping (AAC) improves the model accuracy by exploiting accurate activation information from the full-precision model. Adapti…
Dankook University（檀国大学）；Complutense University of Madrid（马德里康普顿斯大学）
This paper proposes a training method having multiple cyclic training for achieving enhanced performance in low-bit quantized convolutional neural networks (CNNs). Quantization is a popular method for obtaining lightweight CNNs, where the initialization with a pretrained model is widely used to overcome degraded performance in low-resolution quantization. However, large quantization errors between real values and their low-bit quantized ones cause difficulties in achieving acceptable performance for complex networks and large datasets. The proposed training method softly delivers the knowledge…
Shanghai Jiao Tong University（上海交通大学）；École Polytechnique Fédérale de Lausanne (EPFL)（洛桑联邦理工学院）
Pooling and unpooling are two essential operations in constructing hierarchical spherical convolutional neural networks (HS-CNNs) for comprehensive feature learning in the spherical domain. Most existing models employ downsampling-based pooling, which will inevitably incur information loss and cannot adapt to different spherical signals and tasks. Besides, the preserved information after pooling cannot be well restored by the subsequent unpooling to characterize the desirable features for a task. In this paper, we propose a novel framework of HS-CNNs with a lifting structure to learn adaptive …
University of Bonn（波恩大学）；KU Leuven（鲁汶大学）；Bosch Center for Artificial Intelligence（博世人工智能中心）；Stanford University（斯坦福大学）；ETH Zürich（苏黎世联邦理工学院）
While state-of-the-art 3D Convolutional Neural Networks (CNN) achieve very good results on action recognition datasets, they are computationally very expensive and require many GFLOPs. While the GFLOPs of a 3D CNN can be decreased by reducing the temporal feature resolution within the network, there is no setting that is optimal for all input clips. In this work, we therefore introduce a differentiable Similarity Guided Sampling (SGS) module, which can be plugged into any existing 3D CNN architecture. SGS empowers 3D CNNs by learning the similarity of temporal features and grouping similar fea…
Deep neural network quantization with adaptive bitwidths has gained increasing attention due to the ease of model deployment on various platforms with different resource budgets. In this paper, we propose a meta-learning approach to achieve this goal. Specifically, we propose MEBQAT, a simple yet effective way of bitwidth-adaptive quantization aware training (QAT) where meta-learning is effectively combined with QAT by redefining meta-learning tasks to incorporate bitwidths. After being deployed on a platform, MEBQAT allows the (meta-)trained model to be quantized to any candidate bitwidth the…
ETH Zurich（苏黎世联邦理工学院）；Singapore Management University（新加坡管理大学）
We investigate the compression of deep neural networks by quantizing their weights and activations into multiple binary bases, known as multi-bit networks (MBNs), which accelerate the inference and reduce the storage for the deployment on low-resource mobile and embedded platforms. We propose Adaptive Loss-aware Quantization (ALQ), a new MBN quantization pipeline that is able to achieve an average bitwidth below one-bit without notable loss in inference accuracy. Unlike previous MBN quantization solutions that train a quantizer by minimizing the error to reconstruct full precision weights, ALQ…
University of California, Irvine（加利福尼亚大学尔湾分校）；Department of Mathematics（数学系）
We propose an adaptive projection-gradient descent-shrinkage-splitting method (APGDSSM) to integrate penalty based channel pruning into quantization-aware training (QAT). APGDSSM concurrently searches weights in both the quantized subspace and the sparse subspace. APGDSSM uses shrinkage operator and a splitting technique to create sparse weights, as well as the Group Lasso penalty to push the weight sparsity into channel sparsity. In addition, we propose a novel complementary transformed l1 penalty to stabilize the training for extreme compression.
Indian Institute of Information Technology, Allahabad（阿拉哈巴德印度信息技术学院）；Techno India University（印度科技大学）；Indian Statistical Institute（印度统计研究所）
The stochastic gradient descent (SGD) optimizers are generally used to train the convolutional neural networks (CNNs). In recent years, several adaptive momentum based SGD optimizers have been introduced, such as Adam, diffGrad, Radam and AdaBelief. However, the existing SGD optimizers do not exploit the gradient norm of past iterations and lead to poor convergence and performance. In this paper, we propose a novel AdaNorm based SGD optimizers by correcting the norm of gradient in each iteration based on the adaptive training history of gradient norm. By doing so, the proposed optimizers are a…
This paper concerns the use of objectness measures to improve the calibration performance of Convolutional Neural Networks (CNNs). CNNs have proven to be very good classifiers and generally localize objects well; however, the loss functions typically used to train classification CNNs do not penalize inability to localize an object, nor do they take into account an object's relative size in the given image. During training on ImageNet-1K almost all approaches use random crops on the images and this transformation sometimes provides the CNN with background only samples. This causes the classifie…
Neural network models are resource hungry. It is difficult to deploy such deep networks on devices with limited resources, like smart wearables, cellphones, drones, and autonomous vehicles. Low bit quantization such as binary and ternary quantization is a common approach to alleviate this resource requirements. Ternary quantization provides a more flexible model and outperforms binary quantization in terms of accuracy, however doubles the memory footprint and increases the computational cost. Contrary to these approaches, mixed quantized models allow a trade-off between accuracy and memory foo…
To accelerate deep CNN models, this paper proposes a novel spatially adaptive framework that can dynamically generate pixel-wise sparsity according to the input image. The sparse scheme is pixel-wise refined, regional adaptive under a unified importance map, which makes it friendly to hardware implementation. A sparse controlling method is further presented to enable online adjustment for applications with different precision/latency requirements. The sparse model is applicable to a wide range of vision tasks. Experimental results show that this method efficiently improve the computing efficie…
McMaster University（麦克马斯特大学）；Department of Electrical and Computer Engineering（电气与计算机工程系）
Single Image Super-Resolution (SISR) task refers to learn a mapping from low-resolution images to the corresponding high-resolution ones. This task is known to be extremely difficult since it is an ill-posed problem. Recently, Convolutional Neural Networks (CNNs) have achieved state of the art performance on SISR. However, the images produced by CNNs do not contain fine details of the images. Generative Adversarial Networks (GANs) aim to solve this issue and recover sharp details. Nevertheless, GANs are notoriously difficult to train. Besides that, they generate artifacts in the high-resolutio…
Temporal-adaptive Weight Quantization for Spiking Neural Networks
Faculty of Computing, Harbin Institute of Technology（哈尔滨工业大学计算机学院）；Harbin Institute of Technology（哈尔滨工业大学）；Pengcheng Laboratory（鹏城实验室）；Institute of Computing and Intelligence (ICI), Harbin Institute of Technology（哈尔滨工业大学计算机与智能信息学院）；Institute of Automation, Chinese Academy of Sciences（中国科学院自动化研究所）
Recent convolutional neural network (CNN) development continues to advance the state-of-the-art model accuracy for various applications. However, the enhanced accuracy comes at the cost of substantial memory bandwidth and storage requirements and demanding computational resources. Although in the past the quantization methods have effectively reduced the deployment cost for edge devices, it suffers from significant information loss when processing the biased activations of contemporary CNNs. In this paper, we hence introduce an adaptive high-performance quantization method to resolve the issue…
Quant Experts: Token-aware Adaptive Error Reconstruction with Mixture of Experts for Large Vision-Language Models Quantization
State Key Laboratory of Human-Machine Hybrid Augmented Intelligence（人机混合增强智能国家重点实验室）；Institute of Artificial Intelligence and Robotics（人工智能与机器人研究院）；Xi’an Jiaotong University（西安交通大学）
Building footprint extraction in remote sensing data benefits many important applications, such as urban planning and population estimation. Recently, rapid development of Convolutional Neural Networks (CNNs) and open-sourced high resolution satellite building image datasets have pushed the performance boundary further for automated building extractions. However, CNN approaches often generate imprecise building morphologies including noisy edges and round corners. In this paper, we leverage the performance of CNNs, and propose a module that uses prior knowledge of building corners to create an…
Deep neural networks with adaptive configurations have gained increasing attention due to the instant and flexible deployment of these models on platforms with different resource budgets. In this paper, we investigate a novel option to achieve this goal by enabling adaptive bit-widths of weights and activations in the model. We first examine the benefits and challenges of training quantized model with adaptive bit-widths, and then experiment with several approaches including direct adaptation, progressive training and joint training. We discover that joint training is able to produce comparabl…
Huazhong University of Science and Technology（华中科技大学）；Microsoft Research & AI（微软研究院与人工智能部门）；Zhejiang University（浙江大学）；IBM & AI Foundations Learning（IBM人工智能基础学习部门）
Crowd counting is a challenging task due to the large variations in crowd distributions. Previous methods tend to tackle the whole image with a single fixed structure, which is unable to handle diverse complicated scenes with different crowd densities. Hence, we propose the Adaptive Capacity Multi-scale convolutional neural networks (ACM-CNN), a novel crowd counting approach which can assign different capacities to different portions of the input. The intuition is that the model should focus on important regions of the input image and optimize its capacity allocation conditioning on the crowd …
Nowadays most research in visual recognition using Convolutional Neural Networks (CNNs) follows the "deeper model with deeper confidence" belief to gain a higher recognition accuracy. At the same time, deeper model brings heavier computation. On the other hand, for a large chunk of recognition challenges, a system can classify images correctly using simple models or so-called shallow networks. Moreover, the implementation of CNNs faces with the size, weight, and energy constraints on the embedded devices. In this paper, we implement the adaptive switching between shallow and deep networks to r…
AdaLoRA-QAT: Adaptive Low-Rank and Quantization-Aware Segmentation
International Institute of Information Technology (IIIT-H), Hyderabad, India（国际信息技术研究所（IIIT-H），印度海得拉巴）；Nizam’s Institute of Medical Sciences (NIMS), Hyderabad, India（尼扎姆医学科学研究所（NIMS），印度海得拉巴）；The Alan Turing Institute, London（艾伦·图灵研究所，伦敦）；University College London, United Kingdom（伦敦大学学院，英国）
McGill（麦吉尔大学）；Mila（米拉研究所）；Facebook AI Research（Facebook人工智能研究院）；University of Montreal（蒙特利尔大学）；ElementAI（ElementAI公司）
We introduce and study the problem of Online Continual Compression, where one attempts to simultaneously learn to compress and store a representative dataset from a non i.i.d data stream, while only observing each sample once. A naive application of auto-encoders in this setting encounters a major challenge: representations derived from earlier encoder states must be usable by later decoder states. We show how to use discrete auto-encoders to effectively address this challenge and introduce Adaptive Quantization Modules (AQM) to control variation in the compression ability of the module at any…
Precision Neural Network Quantization via Learnable Adaptive Modules
College of Computer Science, Sichuan University（四川大学计算机学院）；Engineering Research Center of Machine Learning and Industry Intelligence, Ministry of Education（机器学习与工业智能教育部工程研究中心）
