In this work, we propose a low-bit training framework for convolutional neural networks, which is built around a novel multi-level scaling (MLS) tensor format. Our framework focuses on reducing the energy consumption of convolution operations by quantizing all the convolution operands to low bit-width format. Specifically, we propose the MLS tensor format, in which the element-wise bit-width can be largely reduced. Then, we describe the dynamic quantization and the low-bit tensor convolution arithmetic to leverage the MLS tensor format efficiently. Experiments show that our framework achieves a superior trade-off between the accuracy and the bit-width than previous low-bit training frameworks. For training a variety of models on CIFAR-10, using 1-bit mantissa and 2-bit exponent is adequate to keep the accuracy loss within $1\%$. And on larger datasets like ImageNet, using 4-bit mantissa and 2-bit exponent is adequate to keep the accuracy loss within $1\%$. Through the energy consumption simulation of the computing units, we can estimate that training a variety of models with our framework could achieve $8.3\sim10.2\times$ and $1.9\sim2.3\times$ higher energy efficiency than training with full-precision and 8-bit floating-point arithmetic, respectively.
Emerging resistive random-access memory (ReRAM) has recently been intensively investigated to accelerate the processing of deep neural networks (DNNs). Due to the in-situ computation capability, analog ReRAM crossbars yield significant throughput improvement and energy reduction compared to traditional digital methods. However, the power hungry analog-to-digital converters (ADCs) prevent the practical deployment of ReRAM-based DNN accelerators on end devices with limited chip area and power budget. We observe that due to the limited bit-density of ReRAM cells, DNN weights are bit sliced and co…
Preferred Networks（优选网络公司）；The University of Tokyo（东京大学）
Tensor decomposition methods are widely used for model compression and fast inference in convolutional neural networks (CNNs). Although many decompositions are conceivable, only CP decomposition and a few others have been applied in practice, and no extensive comparisons have been made between available methods. Previous studies have not determined how many decompositions are available, nor which of them is optimal. In this study, we first characterize a decomposition class specific to CNNs by adopting a flexible graphical notation. The class includes such well-known CNN modules as depthwise s…
We propose DoReFa-Net, a method to train convolutional neural networks that have low bitwidth weights and activations using low bitwidth parameter gradients. In particular, during backward pass, parameter gradients are stochastically quantized to low bitwidth numbers before being propagated to convolutional layers. As convolutions during forward/backward passes can now operate on low bitwidth weights and activations/gradients respectively, DoReFa-Net can use bit convolution kernels to accelerate both training and inference. Moreover, as bit convolutions can be efficiently implemented on CPU, F…
Fully convolutional neural networks give accurate, per-pixel prediction for input images and have applications like semantic segmentation. However, a typical FCN usually requires lots of floating point computation and large run-time memory, which effectively limits its usability. We propose a method to train Bit Fully Convolution Network (BFCN), a fully convolutional neural network that has low bit-width weights and activations. Because most of its computation-intensive convolutions are accomplished between low bit-width numbers, a BFCN can be accelerated by an efficient bit-convolution implem…
University of California, Los Angeles（加州大学洛杉矶分校）；University of California, Irvine（加州大学尔湾分校）
We present LBW-Net, an efficient optimization based method for quantization and training of the low bit-width convolutional neural networks (CNNs). Specifically, we quantize the weights to zero or powers of two by minimizing the Euclidean distance between full-precision weights and quantized weights during backpropagation. We characterize the combinatorial nature of the low bit-width quantization problem. For 2-bit (ternary) CNNs, the quantization of $N$ weights can be done by an exact formula in $O(N\log N)$ complexity. When the bit-width is three and above, we further propose a semi-analytic…
Sparsity helps reduce the computational complexity of deep neural networks by skipping zeros. Taking advantage of sparsity is listed as a high priority in next generation DNN accelerators such as TPU. The structure of sparsity, i.e., the granularity of pruning, affects the efficiency of hardware accelerator design as well as the prediction accuracy. Coarse-grained pruning creates regular sparsity patterns, making it more amenable for hardware acceleration but more challenging to maintain the same accuracy. In this paper we quantitatively measure the trade-off between sparsity regularity and pr…
Mixed-precision quantization can potentially achieve the optimal tradeoff between performance and compression rate of deep neural networks, and thus, have been widely investigated. However, it lacks a systematic method to determine the exact quantization scheme. Previous methods either examine only a small manually-designed search space or utilize a cumbersome neural architecture search to explore the vast search space. These approaches cannot lead to an optimal quantization scheme efficiently. This work proposes bit-level sparsity quantization (BSQ) to tackle the mixed-precision quantization …
We present an overview of techniques for quantizing convolutional neural networks for inference with integer weights and activations. Per-channel quantization of weights and per-layer quantization of activations to 8-bits of precision post-training produces classification accuracies within 2% of floating point networks for a wide variety of CNN architectures. Model sizes can be reduced by a factor of 4 by quantizing weights to 8-bits, even when 8-bit arithmetic is not supported. This can be achieved with simple, post training quantization of weights.We benchmark latencies of quantized networks…
Trinity College Dublin, The University of Dublin（都柏林大学圣三一学院）；The School of Computer Science and Statistics（计算机科学与统计学院）；Institute of Technology Carlow（卡洛理工学院）
Convolutional neural networks (CNNs) are typically trained using 16- or 32-bit floating-point (FP) and researchers show that low-precision floating-point (FP) can be highly effective for inference. Low-precision FP can be implemented in field programmable gate array (FPGA) and application-specific integrated circuit (ASIC) accelerators, but existing processors do not generally support custom precision FP. We propose hardware optimized bitslice-parallel floating-point operators (HOBFLOPS), a method of generating efficient custom-precision emulated bitslice-parallel software FP arithmetic. We ge…
Integrated Systems Laboratory, ETH Zurich（苏黎世联邦理工学院集成系统实验室）
After the tremendous success of convolutional neural networks in image classification, object detection, speech recognition, etc., there is now rising demand for deployment of these compute-intensive ML models on tightly power constrained embedded and mobile systems at low cost as well as for pushing the throughput in data centers. This has triggered a wave of research towards specialized hardware accelerators. Their performance is often constrained by I/O bandwidth and the energy consumption is dominated by I/O transfers to off-chip memory. We introduce and evaluate a novel, hardware-friendly…
Convolutional neural networks have recently achieved significant breakthroughs in various image classification tasks. However, they are computationally expensive,which can make their feasible mplementation on embedded and low-power devices difficult. In this paper convolutional neural network binarization is implemented on GPU-based platforms for real-time inference on resource constrained devices. In binarized networks, all weights and intermediate computations between layers are quantized to +1 and -1, allowing multiplications and additions to be replaced with bit-wise operations between 32-…
Known for their capacity-achieving abilities, polar codes have been selected as the control channel coding scheme for 5G communications. To satisfy the needs of high throughput and low latency, belief propagation (BP) is chosen as the decoding algorithm. However, in general, the error performance of BP is worse than that of enhanced successive cancellation (SC). Recently, critical-set bit-flipping (CS-BF) is applied to BP decoding to lower the error rate. However, its trial and error process result in even longer latency. In this work, we propose a convolutional neural network-assisted bit-fli…
We study channel number reduction in combination with weight binarization (1-bit weight precision) to trim a convolutional neural network for a keyword spotting (classification) task. We adopt a group-wise splitting method based on the group Lasso penalty to achieve over 50% channel sparsity while maintaining the network performance within 0.25% accuracy loss. We show an effective three-stage procedure to balance accuracy and sparsity in network training.
In recent years, the research community has discovered that deep neural networks (DNNs) and convolutional neural networks (CNNs) can yield higher accuracy than all previous solutions to a broad array of machine learning problems. To our knowledge, there is no single CNN/DNN architecture that solves all problems optimally. Instead, the "right" CNN/DNN architecture varies depending on the application at hand. CNN/DNNs comprise an enormous design space. Quantitatively, we find that a small region of the CNN design space contains 30 billion different CNN architectures. In this dissertation, we dev…
The past few years have witnessed growth in the computational requirements for training deep convolutional neural networks. Current approaches parallelize training onto multiple devices by applying a single parallelization strategy (e.g., data or model parallelism) to all layers in a network. Although easy to reason about, these approaches result in suboptimal runtime performance in large-scale distributed training, since different layers in a network may prefer different parallelization strategies. In this paper, we propose layer-wise parallelism that allows each layer in a network to use an …
Recent advances in convolutional neural networks have considered model complexity and hardware efficiency to enable deployment onto embedded systems and mobile devices. For example, it is now well-known that the arithmetic operations of deep networks can be encoded down to 8-bit fixed-point without significant deterioration in performance. However, further reduction in precision down to as low as 3-bit fixed-point results in significant losses in performance. In this paper we propose a new data representation that enables state-of-the-art networks to be encoded to 3 bits with negligible loss i…
University of Toronto（多伦多大学）；Vector Institute（向量研究所）
Triangular, overlapping Mel-scaled filters ("f-banks") are the current standard input for acoustic models that exploit their input's time-frequency geometry, because they provide a psycho-acoustically motivated time-frequency geometry for a speech signal. F-bank coefficients are provably robust to small deformations in the scale. In this paper, we explore two ways in which filter banks can be adjusted for the purposes of speech recognition. First, triangular filters can be replaced with Gabor filters, a compactly supported filter that better localizes events in time, or Gammatone filters, a ps…
Birzeit University（比尔泽特大学）；University College London（伦敦大学学院）；iSize Ltd.（iSize有限公司）
Advanced video classification systems decode video frames to derive the necessary texture and motion representations for ingestion and analysis by spatio-temporal deep convolutional neural networks (CNNs). However, when considering visual Internet-of-Things applications, surveillance systems and semantic crawlers of large video repositories, the video capture and the CNN-based semantic analysis parts do not tend to be co-located. This necessitates the transport of compressed video over networks and incurs significant overhead in bandwidth and energy consumption, thereby significantly undermini…
University of Southampton（南安普顿大学）；Fuzhou University（福州大学）
The traditional image compressors, e.g., BPG and H.266, have achieved great image and video compression quality. Recently, Convolutional Neural Network has been used widely in image compression. We proposed an attention-based convolutional neural network for low bit-rate compression to post-process the output of traditional image compression decoder. Across the experimental results on validation sets, the post-processing module trained by MAE and MS-SSIM losses yields the highest PSNR of 32.10 on average at the bit-rate of 0.15.
University of South Australia（南澳大学）；University of California, San Diego（加州大学圣地亚哥分校）；Western Sydney University（西悉尼大学）
Batch-normalization (BN) layers are thought to be an integrally important layer type in today's state-of-the-art deep convolutional neural networks for computer vision tasks such as classification and detection. However, BN layers introduce complexity and computational overheads that are highly undesirable for training and/or inference on low-power custom hardware implementations of real-time embedded vision systems such as UAVs, robots and Internet of Things (IoT) devices. They are also problematic when batch sizes need to be very small during training, and innovations such as residual connec…
A counter-intuitive property of convolutional neural networks (CNNs) is their inherent susceptibility to adversarial examples, which severely hinders the application of CNNs in security-critical fields. Adversarial examples are similar to original examples but contain malicious perturbations. Adversarial training is a simple and effective defense method to improve the robustness of CNNs to adversarial examples. The mechanisms behind adversarial examples and adversarial training are worth exploring. Therefore, this work investigates similarities and differences between normally trained CNNs (NT…
Understanding the features learned by deep models is important from a model trust perspective, especially as deep systems are deployed in the real world. Most recent approaches for deep feature understanding or model explanation focus on highlighting input data features that are relevant for classification decisions. In this work, we instead take the perspective of relating deep features to well-studied, hand-crafted features that are meaningful for the application of interest. We propose a methodology and set of systematic experiments for exploring deep features in this setting, where input f…
Institute of Computing Technology, Chinese Academy of Sciences（中国科学院计算技术研究所）；University of Chinese Academy of Sciences（中国科学院大学）
Inference efficiency is the predominant consideration in designing deep learning accelerators. Previous work mainly focuses on skipping zero values to deal with remarkable ineffectual computation, while zero bits in non-zero values, as another major source of ineffectual computation, is often ignored. The reason lies on the difficulty of extracting essential bits during operating multiply-and-accumulate (MAC) in the processing element. Based on the fact that zero bits occupy as high as 68.9% fraction in the overall weights of modern deep convolutional neural network models, this paper firstly …
Recently low-bit (e.g., 8-bit) network quantization has been extensively studied to accelerate the inference. Besides inference, low-bit training with quantized gradients can further bring more considerable acceleration, since the backward process is often computation-intensive. Unfortunately, the inappropriate quantization of backward propagation usually makes the training unstable and even crash. There lacks a successful unified low-bit training framework that can support diverse networks on various tasks. In this paper, we give an attempt to build a unified 8-bit (INT8) training framework f…
Stanford University（斯坦福大学）；University of New South Wales（新南威尔士大学）
In this paper, we compress convolutional neural network (CNN) weights post-training via transform quantization. Previous CNN quantization techniques tend to ignore the joint statistics of weights and activations, producing sub-optimal CNN performance at a given quantization bit-rate, or consider their joint statistics during training only and do not facilitate efficient compression of already trained CNN models. We optimally transform (decorrelate) and quantize the weights post-training using a rate-distortion framework to improve compression at any given quantization bit-rate. Transform quant…
Cardiovascular disease (CVDs) is one of the universal deadly diseases, and the detection of it in the early stage is a challenging task to tackle. Recently, deep learning and convolutional neural networks have been employed widely for the classification of objects. Moreover, it is promising that lots of networks can be deployed on wearable devices. An increasing number of methods can be used to realize ECG signal classification for the sake of arrhythmia detection. However, the existing neural networks proposed for arrhythmia detection are not hardware-friendly enough due to a remarkable quant…
Deep neural network is difficult to train and this predicament becomes worse as the depth increases. The essence of this problem exists in the magnitude of backpropagated errors that will result in gradient vanishing or exploding phenomenon. We show that a variant of regularizer which utilizes orthonormality among different filter banks can alleviate this problem. Moreover, we design a backward error modulation mechanism based on the quasi-isometry assumption between two consecutive parametric layers. Equipped with these two ingredients, we propose several novel optimization solutions that can…
Exploring Layer-wise Information Effectiveness for Post-Training Quantization in Small Language Models
The University of Hong Kong（香港大学）；Huazhong University of Science and Technology（华中科技大学）；Shenzhen International Graduate School, Tsinghua University（清华大学深圳国际研究生学院）
School of Computing, Newcastle University（纽卡斯尔大学计算机学院）；Department of Computer Science, Durham University（杜伦大学计算机科学系）
With the growing significance of graphs as an effective representation of data in numerous applications, efficient graph analysis using modern machine learning is receiving a growing level of attention. Deep learning approaches often operate over the entire adjacency matrix -- as the input and intermediate network layers are all designed in proportion to the size of the adjacency matrix -- leading to intensive computation and large memory requirements as the graph size increases. It is therefore desirable to identify efficient measures to reduce both run-time and memory requirements allowing f…
Asia University（亚洲大学）；China Medical University Hospital, China Medical University（中国医药大学附设医院，中国医药大学）；Universitas Muhammadiyah Yogyakarta（穆罕默迪亚日惹大学）；Western Norway University of Applied Sciences（挪威西部应用科学大学）
The spread of PM2.5 pollutants that endanger health is difficult to predict because it involves many atmospheric variables. These micron particles can spread rapidly from their source to residential areas, increasing the risk of respiratory disease if exposed for long periods. The prediction system of PM2.5 propagation provides more detailed and accurate information as an early warning system to reduce health impacts on the community. According to the idea of transformative computing, the approach we propose in this paper allows computation on the dataset obtained from massive-scale PM2.5 sens…
Loom (LM), a hardware inference accelerator for Convolutional Neural Networks (CNNs) is presented. In LM every bit of data precision that can be saved translates to proportional performance gains. Specifically, for convolutional layers LM's execution time scales inversely proportionally with the precisions of both weights and activations. For fully-connected layers LM's performance scales inversely proportionally with the precision of the weights. LM targets area- and bandwidth-constrained System-on-a-Chip designs such as those found on mobile devices that cannot afford the multi-megabyte buff…
Speech coding facilitates the transmission of speech over low-bandwidth networks with minimal distortion. Neural-network based speech codecs have recently demonstrated significant improvements in quality over traditional approaches. While this new generation of codecs is capable of synthesizing high-fidelity speech, their use of recurrent or convolutional layers often restricts their effective receptive fields, which prevents them from compressing speech efficiently. We propose to further reduce the bitrate of neural speech codecs through the use of pretrained Transformers, capable of exploiti…
In recent years increasingly complex architectures for deep convolution networks (DCNs) have been proposed to boost the performance on image recognition tasks. However, the gains in performance have come at a cost of substantial increase in computation and model storage resources. Fixed point implementation of DCNs has the potential to alleviate some of these complexities and facilitate potential deployment on embedded hardware. In this paper, we propose a quantizer design for fixed point implementation of DCNs. We formulate and solve an optimization problem to identify optimal fixed point bit…
Quantization Meets Reasoning: Exploring and Mitigating Degradation of Low-Bit LLMs in Mathematical Reasoning
The Hong Kong Polytechnic University（香港理工大学）；University of California, Santa Barbara（加州大学圣芭芭拉分校）；Peking University（北京大学）；The Hong Kong University of Science and Technology（香港科学与技术大学）；The University of Hong Kong（香港大学）
Exploring specialization and sensitivity of convolutional neural networks in the context of simultaneous image augmentations
Marchuk Institute of Numerical Mathematics（马尔库克数值数学研究所）；Skolkovo Institute of Science and Technology（斯克尔科沃科学与技术研究所）；Faculty of Computational Mathematics and Cybernetics, Lomonosov Moscow State University（洛蒙诺索夫莫斯科国立大学计算数学与 cybernetics 学院）；Artificial Intelligence Research Institute (AIRI)（人工智能研究 institute (AIRI)）
Herein, a bit-wise Convolutional Neural Network (CNN) in-memory accelerator is implemented using Spin-Orbit Torque Magnetic Random Access Memory (SOT-MRAM) computational sub-arrays. It utilizes a novel AND-Accumulation method capable of significantly-reduced energy consumption within convolutional layers and performs various low bit-width CNN inference operations entirely within MRAM. Power-intermittence resiliency is also enhanced by retaining the partial state information needed to maintain computational forward-progress, which is advantageous for battery-less IoT nodes. Simulation results i…
Veermata Jijabai Technological Institute (VJTI)（维尔马塔·吉贾巴伊技术学院（VJTI））
In recent years, neural network-based image compression techniques have been able to outperform traditional codecs and have opened the gates for the development of learning-based video codecs. However, to take advantage of the high temporal correlation in videos, more sophisticated architectures need to be employed. This paper presents PredEncoder, a hybrid video compression framework based on the concept of predictive auto-encoding that models the temporal correlations between consecutive video frames using a prediction network which is then combined with a progressive encoder network to expl…
Singapore Management University（新加坡管理大学）；The Open University, UK（英国开放大学）
Towards the vision of translating code that implements an algorithm from one programming language into another, this paper proposes an approach for automated program classification using bilateral tree-based convolutional neural networks (BiTBCNNs). It is layered on top of two tree-based convolutional neural networks (TBCNNs), each of which recognizes the algorithm of code written in an individual programming language. The combination layer of the networks recognizes the similarities and differences among code in different programming languages. The BiTBCNNs are trained using the source code i…
University of Missouri（密苏里大学）；The University of Texas at Dallas（德克萨斯大学达拉斯分校）
Systolic array-based deep neural network (DNN) accelerators have recently gained prominence for their low computational cost. However, their high energy consumption poses a bottleneck to their deployment in energy-constrained devices. To address this problem, approximate computing can be employed at the cost of some tolerable accuracy loss. However, such small accuracy variations may increase the sensitivity of DNNs towards undesired subtle disturbances, such as permanent faults. The impact of permanent faults in accurate DNNs has been thoroughly investigated in the literature. Conversely, the…
There is a need for automatic diagnosis of certain diseases from medical images that could help medical practitioners for further assessment towards treating the illness. Alzheimers disease is a good example of a disease that is often misdiagnosed. Alzheimers disease (Hear after referred to as AD), is caused by atrophy of certain brain regions and by brain cell death and is the leading cause of dementia and memory loss [1]. MRI scans reveal this information but atrophied regions are different for different individuals which makes the diagnosis a bit more trickier and often gets misdiagnosed [1…
Efficient point cloud compression is fundamental to enable the deployment of virtual and mixed reality applications, since the number of points to code can range in the order of millions. In this paper, we present a novel data-driven geometry compression method for static point clouds based on learned convolutional transforms and uniform quantization. We perform joint optimization of both rate and distortion using a trade-off parameter. In addition, we cast the decoding process as a binary classification of the point cloud occupancy map. Our method outperforms the MPEG reference solution in te…
We introduce a novel convolutional network architecture with an interpretable latent space for modeling guitar amplifiers. Leveraging domain knowledge of popular amplifiers spanning a range of styles, the proposed system intuitively combines or subtracts characteristics of different amplifiers, allowing musicians to design entirely new guitar timbres.
Deep Neural Networks (DNNs) and Convolutional Neural Networks (CNNs) are useful for many practical tasks in machine learning. Synaptic weights, as well as neuron activation functions within the deep network are typically stored with high-precision formats, e.g. 32 bit floating point. However, since storage capacity is limited and each memory access consumes power, both storage capacity and memory access are two crucial factors in these networks. Here we present a method and present the ADaPTION toolbox to extend the popular deep learning library Caffe to support training of deep CNNs with redu…
We propose and study a new projection formula for training binary weight convolutional neural networks. The projection formula measures the error in approximating a full precision (32 bit) vector by a 1-bit vector in the l_1 norm instead of the standard l_2 norm. The l_1 projector is in closed analytical form and involves a median computation instead of an arithmatic average in the l_2 projector. Experiments on 10 keywords classification show that the l_1 (median) BinaryConnect (BC) method outperforms the regular BC, regardless of cold or warm start. The binary network trained by median BC and…
Exploring the Role of Convolutional Neural Networks (CNN) in Dental Radiography Segmentation: A Comprehensive Systematic Literature Review
Research Groups in Intelligent Machines (REGIM Lab), University of Sfax, National Engineering School of Sfax (ENIS)（智能机器研究组（REGIM实验室），斯法克斯大学，斯法克斯国家工程学院）
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…
MIT-IBM Watson AI Lab（麻省理工学院-IBM沃森人工智能实验室）；IBM Research（IBM研究院）；IBM Research AI（IBM人工智能研究院）；Elliptic（Elliptic公司）；MIT CSAIL（麻省理工学院计算机科学与人工智能实验室）
Anti-money laundering (AML) regulations play a critical role in safeguarding financial systems, but bear high costs for institutions and drive financial exclusion for those on the socioeconomic and international margins. The advent of cryptocurrency has introduced an intriguing paradox: pseudonymity allows criminals to hide in plain sight, but open data gives more power to investigators and enables the crowdsourcing of forensic analysis. Meanwhile advances in learning algorithms show great promise for the AML toolkit. In this workshop tutorial, we motivate the opportunity to reconcile the caus…
Artificial intelligence (AI) is playing an increasingly significant role in our everyday lives. This trend is expected to continue, especially with recent pushes to move more AI to the edge. However, one of the biggest challenges associated with AI on edge devices (mobile phones, unmanned vehicles, sensors, etc.) is their associated size, weight, and power constraints. In this work, we consider the scenario where an AI system may need to operate at less-than-maximum accuracy in order to meet application-dependent energy requirements. We propose a simple function that divides the cost of using …
