We tackle the problem of unsupervised visual descriptors compression, which is a key ingredient of large-scale image retrieval systems. While the deep learning machinery has benefited literally all computer vision pipelines, the existing state-of-the-art compression methods employ shallow architectures, and we aim to close this gap by our paper. In more detail, we introduce a DNN architecture for the unsupervised compressed-domain retrieval, based on multi-codebook quantization. The proposed architecture is designed to incorporate both fast data encoding and efficient distances computation via lookup tables. We demonstrate the exceptional advantage of our scheme over existing quantization approaches on several datasets of visual descriptors via outperforming the previous state-of-the-art by a large margin.
University of Waterloo（滑铁卢大学）；Vector Institute（向量研究所）
An effective unsupervised hashing algorithm leads to compact binary codes preserving the neighborhood structure of data as much as possible. One of the most established schemes for unsupervised hashing is to reduce the dimensionality of data and then find a rigid (neighbourhood-preserving) transformation that reduces the quantization error. Although employing rigid transformations is effective, we may not reduce quantization loss to the ultimate limits. As well, reducing dimensionality and quantization loss in two separate steps seems to be sub-optimal. Motivated by these shortcomings, we prop…
Shanghai Jiao Tong University（上海交通大学）；South University of Science and Technology（南方科技大学）；University of California, Riverside（加州大学河滨分校）
As it requires a huge number of parameters when exposed to high dimensional inputs in video detection and classification, there is a grand challenge to develop a compact yet accurate video comprehension at terminal devices. Current works focus on optimizations of video detection and classification in a separated fashion. In this paper, we introduce a video comprehension (object detection and action recognition) system for terminal devices, namely DEEPEYE. Based on You Only Look Once (YOLO), we have developed an 8-bit quantization method when training YOLO; and also developed a tensorized-compr…
University of Catania（卡塔尼亚大学）；University of Cagliari（卡利亚里大学）
The exploitation of traces in JPEG double compressed images is of utter importance for investigations. Properly exploiting such insights, First Quantization Estimation (FQE) could be performed in order to obtain source camera model identification (CMI) and therefore reconstruct the history of a digital image. In this paper, a method able to estimate the first quantization factors for JPEG double compressed images is presented, employing a mixed statistical and Machine Learning approach. The presented solution is demonstrated to work without any a-priori assumptions about the quantization matri…
We propose the position-based scaled gradient (PSG) that scales the gradient depending on the position of a weight vector to make it more compression-friendly. First, we theoretically show that applying PSG to the standard gradient descent (GD), which is called PSGD, is equivalent to the GD in the warped weight space, a space made by warping the original weight space via an appropriately designed invertible function. Second, we empirically show that PSG acting as a regularizer to a weight vector is favorable for model compression domains such as quantization and pruning. PSG reduces the gap be…
This paper shows how to reduce the computational cost for a variety of common machine vision tasks by operating directly in the compressed domain, particularly in the context of hardware acceleration. Pyramid Vector Quantization (PVQ) is the compression technique of choice and its properties are exploited to simplify Support Vector Machines (SVM), Convolutional Neural Networks(CNNs), Histogram of Oriented Gradients (HOG) features, interest points matching and other algorithms.
Luminance-Aware Statistical Quantization: Unsupervised Hierarchical Learning for Illumination Enhancement
College of Electronic Science and Technology, National University of Defense Technology（电子科学与技术学院，国防科技大学）；College of Electronic and Information Engineering, Beihang University（电子与信息工程学院，北航）
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…
McGill University（麦吉尔大学）；City University of Hong Kong（香港城市大学）
Model quantization enables the deployment of deep neural networks under resource-constrained devices. Vector quantization aims at reducing the model size by indexing model weights with full-precision embeddings, i.e., codewords, while the index needs to be restored to 32-bit during computation. Binary and other low-precision quantization methods can reduce the model size up to 32$\times$, however, at the cost of a considerable accuracy drop. In this paper, we propose an efficient framework for ternary quantization to produce smaller and more accurate compressed models. By integrating hypersphe…
All Lossy compression algorithms employ similar compression schemes -- frequency domain transform followed by quantization and lossless encoding schemes. They target tradeoffs by quantizating high frequency data to increase compression rates which come at the cost of higher image distortion. We propose a new method of optimizing quantization tables using Deep Learning and a minimax loss function that more accurately measures the tradeoffs between rate and distortion parameters (RD) than previous methods. We design a convolutional neural network (CNN) that learns a mapping between image blocks …
Facebook AI Research（Facebook人工智能研究院）；Univ Rennes（雷恩大学）；Inria（法国国家信息与自动化研究所）；CNRS（法国国家科学研究中心）；IRISA（IRISA研究所）
In this paper, we address the problem of reducing the memory footprint of convolutional network architectures. We introduce a vector quantization method that aims at preserving the quality of the reconstruction of the network outputs rather than its weights. The principle of our approach is that it minimizes the loss reconstruction error for in-domain inputs. Our method only requires a set of unlabelled data at quantization time and allows for efficient inference on CPU by using byte-aligned codebooks to store the compressed weights. We validate our approach by quantizing a high performing Res…
A Closed-Loop Evaluation of Capability Loss and Recovery in Compressed Driving Policies
Universitas Muhammadiyah Yogyakarta（日惹穆罕马迪亚大学）；Nara Institute of Science and Technology (NAIST)（奈良科学技术研究所）；King Fahd University of Petroleum and Minerals (KFUPM)（法赫德国王石油与矿业大学）
JPEG is one of the most widely used image formats, but in some ways remains surprisingly unoptimized, perhaps because some natural optimizations would go outside the standard that defines JPEG. We show how to improve JPEG compression in a standard-compliant, backward-compatible manner, by finding improved default quantization tables. We describe a simulated annealing technique that has allowed us to find several quantization tables that perform better than the industry standard, in terms of both compressed size and image fidelity. Specifically, we derive tables that reduce the FSIM error by ov…
Artificial Intelligence Research Lab（人工智能研究实验室）；Fit-For-Purpose Technologies
Accelerating deep neural network (DNN) inference on resource-limited devices is one of the most important barriers to ensuring a wider and more inclusive adoption. To alleviate this, DNN binary quantization for faster convolution and memory savings is one of the most promising strategies despite its serious drop in accuracy. The present paper therefore proposes a novel binary quantization function based on quantized compressed sensing (QCS). Theoretical arguments conjecture that our proposal preserves the practical benefits of standard methods, while reducing the quantization error and the res…
Capital Normal University（首都师范大学）；University of Chinese Academy of Sciences（中国科学院大学）；Max-Planck-Institute for Mathematics in the Sciences（马克斯·普朗克数学科学研究所）；Kavli Institute for Theoretical Sciences（卡弗里理论科学研究所）；CAS Center for Excellence in Topological Quantum Computation（中国科学院拓扑量子计算卓越创新中心）；ICFO-Institut de Ciencies Fotoniques（光子科学研究所）；The Barcelona Institute of Science and Technology（巴塞罗那科学技术研究所）；ICREA（加泰罗尼亚高等研究院）
We propose tensor-network compressed sensing (TNCS) by combining the ideas of compressed sensing, tensor network (TN), and machine learning, which permits novel and efficient quantum communications of realistic data. The strategy is to use the unsupervised TN machine learning algorithm to obtain the entangled state $|Ψ\rangle$ that describes the probability distribution of a huge amount of classical information considered to be communicated. To transfer a specific piece of information with $|Ψ\rangle$, our proposal is to encode such information in the separable state with the minimal distance …
Neural Machine Translation (NMT) is resource intensive. We design a quantization procedure to compress NMT models better for devices with limited hardware capability. Because most neural network parameters are near zero, we employ logarithmic quantization in lieu of fixed-point quantization. However, we find bias terms are less amenable to log quantization but note they comprise a tiny fraction of the model, so we leave them uncompressed. We also propose to use an error-feedback mechanism during retraining, to preserve the compressed model as a stale gradient. We empirically show that NMT mode…
Through a Compressed Lens: Investigating The Impact of Quantization on Factual Knowledge Recall
Quality and Usability Lab, Technische Universität Berlin（柏林技术大学质量与可用性实验室）；German Research Center for Artificial Intelligence (DFKI)（德国人工智能研究中心）；Saarland Informatics Campus（萨尔州信息学院）；LMU Munich（慕尼黑大学）；Bosch Center for Artificial Intelligence (BCAI)（博世人工智能中心）；Munich Center for Machine Learning (MCML)（慕尼黑机器学习中心）；Centre for European Research in Trusted AI (CERTAIN)（可信人工智能欧洲研究中心）；BIFOLD – Berlin Institute for the Foundations of Learning and Data（柏林学习与数据基础研究院）；Technical University of Munich（慕尼黑技术大学）
The increasing size of neural network models has been critical for improvements in their accuracy, but device memory is not growing at the same rate. This creates fundamental challenges for training neural networks within limited memory environments. In this work, we propose ActNN, a memory-efficient training framework that stores randomly quantized activations for back propagation. We prove the convergence of ActNN for general network architectures, and we characterize the impact of quantization on the convergence via an exact expression for the gradient variance. Using our theory, we propose…
Indian Institute of Technology Gandhinagar（印度理工学院甘地讷格尔分校）
The double JPEG compression detection has received much attention in recent years due to its applicability as a forensic tool for the most widely used JPEG file format. Existing state-of-the-art CNN-based methods either use histograms of all the frequencies or rely on heuristics to select histograms of specific low frequencies to classify single and double compressed images. However, even amidst lower frequencies of double compressed images/patches, histograms of all the frequencies do not have distinguishable features to separate them from single compressed images. This paper directly extract…
CCAD: Compressed Global Feature Conditioned Anomaly Detection
Columbia University（哥伦比亚大学）；Ping An Property & Casualty Insurance Company（平安财产保险股份有限公司）；University of Michigan（密歇根大学）；University of California at Berkeley（加州大学伯克利分校）；Stevens Institute of Technology（史蒂文斯理工学院）；VisionX LLC.（VisionX有限责任公司）
Neural network compression methods like pruning and quantization are very effective at efficiently deploying Deep Neural Networks (DNNs) on edge devices. However, DNNs remain vulnerable to adversarial examples-inconspicuous inputs that are specifically designed to fool these models. In particular, Universal Adversarial Perturbations (UAPs), are a powerful class of adversarial attacks which create adversarial perturbations that can generalize across a large set of inputs. In this work, we analyze the effect of various compression techniques to UAP attacks, including different forms of pruning a…
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…
The University of Tokyo（东京大学）；Institute for Physics of Intelligence（智能物理研究所）
We consider the general problem of recovering a high-dimensional signal from noisy quantized measurements. Quantization, especially coarse quantization such as 1-bit sign measurements, leads to severe information loss and thus a good prior knowledge of the unknown signal is helpful for accurate recovery. Motivated by the power of score-based generative models (SGM, also known as diffusion models) in capturing the rich structure of natural signals beyond simple sparsity, we propose an unsupervised data-driven approach called quantized compressed sensing with SGM (QCS-SGM), where the prior distr…
University of Maryland, College Park（马里兰大学帕克分校）；University of Maryland Institute for Advanced Computer Studies（马里兰大学高级计算机研究所）
In recent years, neural network approaches have been widely adopted for machine learning tasks, with applications in computer vision. More recently, unsupervised generative models based on neural networks have been successfully applied to model data distributions via low-dimensional latent spaces. In this paper, we use Generative Adversarial Networks (GANs) to impose structure in compressed sensing problems, replacing the usual sparsity constraint. We propose to train the GANs in a task-aware fashion, specifically for reconstruction tasks. We also show that it is possible to train our model wi…
Compressing word embeddings is important for deploying NLP models in memory-constrained settings. However, understanding what makes compressed embeddings perform well on downstream tasks is challenging---existing measures of compression quality often fail to distinguish between embeddings that perform well and those that do not. We thus propose the eigenspace overlap score as a new measure. We relate the eigenspace overlap score to downstream performance by developing generalization bounds for the compressed embeddings in terms of this score, in the context of linear and logistic regression. W…
Rensselaer Polytechnic Institute（伦斯勒理工学院）；IBM Thomas J. Watson Research Center（IBM托马斯·J·沃森研究中心）
We propose Compressed Vertical Federated Learning (C-VFL) for communication-efficient training on vertically partitioned data. In C-VFL, a server and multiple parties collaboratively train a model on their respective features utilizing several local iterations and sharing compressed intermediate results periodically. Our work provides the first theoretical analysis of the effect message compression has on distributed training over vertically partitioned data. We prove convergence of non-convex objectives at a rate of $O(\frac{1}{\sqrt{T}})$ when the compression error is bounded over the course…
Shanghai Jiao Tong University（上海交通大学）；University of Chinese Academy of Sciences（中国科学院大学）
Network quantization is an effective solution to compress deep neural networks for practical usage. Existing network quantization methods cannot sufficiently exploit the depth information to generate low-bit compressed network. In this paper, we propose two novel network quantization approaches, single-level network quantization (SLQ) for high-bit quantization and multi-level network quantization (MLQ) for extremely low-bit quantization (ternary).We are the first to consider the network quantization from both width and depth level. In the width level, parameters are divided into two parts: one…
Deep neural network pruning and quantization techniques have demonstrated it is possible to achieve high levels of compression with surprisingly little degradation to test set accuracy. However, this measure of performance conceals significant differences in how different classes and images are impacted by model compression techniques. We find that models with radically different numbers of weights have comparable top-line performance metrics but diverge considerably in behavior on a narrow subset of the dataset. This small subset of data points, which we term Pruning Identified Exemplars (PIE…
African Institute for Mathematical Sciences（非洲数学科学研究所）
Deep learning approaches have achieved unprecedented performance in visual recognition tasks such as object detection and pose estimation. However, state-of-the-art models have millions of parameters represented as floats which make them computationally expensive and constrain their deployment on hardware such as mobile phones and IoT nodes. Most commonly, activations of deep neural networks tend to be sparse thus proving that models are over parametrized with redundant neurons. Model compression techniques, such as pruning and quantization, have recently shown promising results by improving m…
IIIT Hyderabad（印度海得拉巴国际信息技术学院）；IBM Research - India（IBM印度研究院）
Large number of weights in deep neural networks makes the models difficult to be deployed in low memory environments such as, mobile phones, IOT edge devices as well as "inferencing as a service" environments on cloud. Prior work has considered reduction in the size of the models, through compression techniques like pruning, quantization, Huffman encoding etc. However, efficient inferencing using the compressed models has received little attention, specially with the Huffman encoding in place. In this paper, we propose efficient parallel algorithms for inferencing of single image and batches, …
RotaryQuant: Fitting 120B MoE Models on Consumer Hardware via Fused Compressed-Space Attention
Cognizant AI & Analytics（高知特人工智能与分析部门）；Cognizant AI Lab（高知特人工智能实验室）；Southern Methodist University (SMU)（南卫理公会大学）；University of Maryland, Baltimore County（马里兰大学巴尔的摩县分校）
The popularity and widespread use of pruning and quantization is driven by the severe resource constraints of deploying deep neural networks to environments with strict latency, memory and energy requirements. These techniques achieve high levels of compression with negligible impact on top-line metrics (top-1 and top-5 accuracy). However, overall accuracy hides disproportionately high errors on a small subset of examples; we call this subset Compression Identified Exemplars (CIE). We further establish that for CIE examples, compression amplifies existing algorithmic bias. Pruning disproportio…
