Embedding layers are commonly used to map discrete symbols into continuous embedding vectors that reflect their semantic meanings. Despite their effectiveness, the number of parameters in an embedding layer increases linearly with the number of symbols and poses a critical challenge on memory and storage constraints. In this work, we propose a generic and end-to-end learnable compression framework termed differentiable product quantization (DPQ). We present two instantiations of DPQ that leverage different approximation techniques to enable differentiability in end-to-end learning. Our method can readily serve as a drop-in alternative for any existing embedding layer. Empirically, DPQ offers significant compression ratios (14-238$\times$) at negligible or no performance cost on 10 datasets across three different language tasks.
We propose DiffQ a differentiable method for model compression for quantizing model parameters without gradient approximations (e.g., Straight Through Estimator). We suggest adding independent pseudo quantization noise to model parameters during training to approximate the effect of a quantization operator. DiffQ is differentiable both with respect to the unquantized weights and the number of bits used. Given a single hyper-parameter balancing between the quantized model size and accuracy, DiffQ optimizes the number of bits used per individual weight or groups of weights, in end-to-end trainin…
Qualcomm AI Research（高通人工智能研究院）；University of California, Irvine（加利福尼亚大学尔湾分校）
We present a differentiable joint pruning and quantization (DJPQ) scheme. We frame neural network compression as a joint gradient-based optimization problem, trading off between model pruning and quantization automatically for hardware efficiency. DJPQ incorporates variational information bottleneck based structured pruning and mixed-bit precision quantization into a single differentiable loss function. In contrast to previous works which consider pruning and quantization separately, our method enables users to find the optimal trade-off between both in a single training procedure. To utilize …
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）；Northeastern University（东北大学）；Syracuse University（雪城大学）；Rutgers University（罗格斯大学）
Continuous representation of words is a standard component in deep learning-based NLP models. However, representing a large vocabulary requires significant memory, which can cause problems, particularly on resource-constrained platforms. Therefore, in this paper we propose an isotropic iterative quantization (IIQ) approach for compressing embedding vectors into binary ones, leveraging the iterative quantization technique well established for image retrieval, while satisfying the desired isotropic property of PMI based models. Experiments with pre-trained embeddings (i.e., GloVe and HDC) demons…
With the development of deep neural networks, the size of network models becomes larger and larger. Model compression has become an urgent need for deploying these network models to mobile or embedded devices. Model quantization is a representative model compression technique. Although a lot of quantization methods have been proposed, many of them suffer from a high quantization error caused by a long-tail distribution of network weights. In this paper, we propose a novel quantization method, called weight normalization based quantization (WNQ), for model compression. WNQ adopts weight normali…
The Chinese University of Hong Kong（香港中文大学）；SenseBrain, Ltd（SenseBrain有限公司）；Shanghai AI Lab（上海人工智能实验室）；The University of Hong Kong（香港大学）
Model quantization is challenging due to many tedious hyper-parameters such as precision (bitwidth), dynamic range (minimum and maximum discrete values) and stepsize (interval between discrete values). Unlike prior arts that carefully tune these values, we present a fully differentiable approach to learn all of them, named Differentiable Dynamic Quantization (DDQ), which has several benefits. (1) DDQ is able to quantize challenging lightweight architectures like MobileNets, where different layers prefer different quantization parameters. (2) DDQ is hardware-friendly and can be easily implement…
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…
Large pre-trained language models such as BERT have shown their effectiveness in various natural language processing tasks. However, the huge parameter size makes them difficult to be deployed in real-time applications that require quick inference with limited resources. Existing methods compress BERT into small models while such compression is task-independent, i.e., the same compressed BERT for all different downstream tasks. Motivated by the necessity and benefits of task-oriented BERT compression, we propose a novel compression method, AdaBERT, that leverages differentiable Neural Architec…
Deep neural network (DNN) model compression for efficient on-device inference is becoming increasingly important to reduce memory requirements and keep user data on-device. To this end, we propose a novel differentiable k-means clustering layer (DKM) and its application to train-time weight clustering-based DNN model compression. DKM casts k-means clustering as an attention problem and enables joint optimization of the DNN parameters and clustering centroids. Unlike prior works that rely on additional regularizers and parameters, DKM-based compression keeps the original loss function and model…
Data hiding is one widely used approach for protecting authentication and ownership. Most multimedia content like images and videos are transmitted or saved in the compressed form. This kind of lossy compression, such as JPEG, can destroy the hidden data, which raises the need of robust data hiding. It is still an open challenge to achieve the goal of data hiding that can be against these compressions. Recently, deep learning has shown large success in data hiding, while non-differentiability of JPEG makes it challenging to train a deep pipeline for improving robustness against lossy compressi…
Beihang University（北京航空航天大学）；SenseTime Group Limited（商汤集团有限公司）；Beijing Institute of Technology（北京理工大学）
Hardware-friendly network quantization (e.g., binary/uniform quantization) can efficiently accelerate the inference and meanwhile reduce memory consumption of the deep neural networks, which is crucial for model deployment on resource-limited devices like mobile phones. However, due to the discreteness of low-bit quantization, existing quantization methods often face the unstable training process and severe performance degradation. To address this problem, in this paper we propose Differentiable Soft Quantization (DSQ) to bridge the gap between the full-precision and low-bit networks. DSQ can …
In many learning situations, resources at inference time are significantly more constrained than resources at training time. This paper studies a general paradigm, called Differentiable ARchitecture Compression (DARC), that combines model compression and architecture search to learn models that are resource-efficient at inference time. Given a resource-intensive base architecture, DARC utilizes the training data to learn which sub-components can be replaced by cheaper alternatives. The high-level technique can be applied to any neural architecture, and we report experiments on state-of-the-art…
This paper explores the problem of learning transforms for image compression via autoencoders. Usually, the rate-distortion performances of image compression are tuned by varying the quantization step size. In the case of autoen-coders, this in principle would require learning one transform per rate-distortion point at a given quantization step size. Here, we show that comparable performances can be obtained with a unique learned transform. The different rate-distortion points are then reached by varying the quantization step size at test time. This approach saves a lot of training time.
Hong Kong University of Science and Technology（香港科技大学）；Mohamed bin Zayed University of Artificial Intelligence（穆罕默德·本·扎耶德人工智能大学）；Jockey Club Institute for Advanced Study, HKUST（香港科技大学赛马会高等研究院）；Reality Labs, Meta Inc（Meta公司现实实验室）；ACCESS - AI Chip Center for Emerging Smart Systems（新兴智能系统AI芯片中心（ACCESS））；Carnegie Mellon University（卡内基梅隆大学）
In order to deploy deep models in a computationally efficient manner, model quantization approaches have been frequently used. In addition, as new hardware that supports mixed bitwidth arithmetic operations, recent research on mixed precision quantization (MPQ) begins to fully leverage the capacity of representation by searching optimized bitwidths for different layers and modules in a network. However, previous studies mainly search the MPQ strategy in a costly scheme using reinforcement learning, neural architecture search, etc., or simply utilize partial prior knowledge for bitwidth assignm…
DiVeQ: Differentiable Vector Quantization Using the Reparameterization Trick
ELLIS Institute Finland & Department of Computer Science, Aalto University, Finland（芬兰ELLIS研究所及阿尔托大学计算机科学系）；Department of Information and Communications Engineering, Aalto University, Finland（芬兰阿尔托大学信息与通信工程系）
Columbia University（哥伦比亚大学）；Peking University（北京大学）；Deepwise AI Lab（深度智耀人工智能实验室）；The University of Hong Kong（香港大学）
This paper presents a novel network compression framework Kernel Quantization (KQ), targeting to efficiently convert any pre-trained full-precision convolutional neural network (CNN) model into a low-precision version without significant performance loss. Unlike existing methods struggling with weight bit-length, KQ has the potential in improving the compression ratio by considering the convolution kernel as the quantization unit. Inspired by the evolution from weight pruning to filter pruning, we propose to quantize in both kernel and weight level. Instead of representing each weight paramete…
Deep Neural Networks (DNNs) are applied in a wide range of usecases. There is an increased demand for deploying DNNs on devices that do not have abundant resources such as memory and computation units. Recently, network compression through a variety of techniques such as pruning and quantization have been proposed to reduce the resource requirement. A key parameter that all existing compression techniques are sensitive to is the compression ratio (e.g., pruning sparsity, quantization bitwidth) of each layer. Traditional solutions treat the compression ratios of each layer as hyper-parameters, …
Model compression has gained a lot of attention due to its ability to reduce hardware resource requirements significantly while maintaining accuracy of DNNs. Model compression is especially useful for memory-intensive recurrent neural networks because smaller memory footprint is crucial not only for reducing storage requirement but also for fast inference operations. Quantization is known to be an effective model compression method and researchers are interested in minimizing the number of bits to represent parameters. In this work, we introduce an iterative technique to apply quantization, pr…
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…
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 …
D-QRELO: Training- and Data-Free Delta Compression for Large Language Models via Quantization and Residual Low-Rank Approximation
Harbin Institute of Technology, Shenzhen, China（哈尔滨工业大学（深圳））；TeleAI of China Telecom（中国电信TeleAI）；The Hong Kong Polytechnic University（香港理工大学）；The University of Hong Kong（香港大学）
Model compression techniques, such as pruning and quantization, are becoming increasingly important to reduce the memory footprints and the amount of computations. Despite model size reduction, achieving performance enhancement on devices is, however, still challenging mainly due to the irregular representations of sparse matrix formats. This paper proposes a new weight representation scheme for Sparse Quantized Neural Networks, specifically achieved by fine-grained and unstructured pruning method. The representation is encrypted in a structured regular format, which can be efficiently decoded…
HESTIA: A Hessian-Guided Differentiable Quantization-Aware Training Framework for Extremely Low-Bit LLMs
School of Software and Microelectronics, Peking University, Beijing, China（软件与微电子学院，北京大学，北京，中国）；School of Computer Science, Peking University, Beijing, China（计算机科学学院，北京大学，北京，中国）
Differentiable Earth Mover's Distance for Data Compression at the High-Luminosity LHC
University of California at San Diego（加利福尼亚大学圣迭戈分校）；Fermi National Accelerator Laboratory（费米国家加速器实验室）；European Organization for Nuclear Research (CERN)（欧洲核子研究中心）
Facebook AI Research（Facebook人工智能研究院）；LORIA（洛里亚计算机科学实验室）；Univ Lyon（里昂大学）；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）；ENS de Lyon（里昂高等师范学校）；UCB Lyon 1（里昂第一大学）；LIP UMR 5668（里昂第一大学计算机科学实验室）
We tackle the problem of producing compact models, maximizing their accuracy for a given model size. A standard solution is to train networks with Quantization Aware Training, where the weights are quantized during training and the gradients approximated with the Straight-Through Estimator. In this paper, we extend this approach to work beyond int8 fixed-point quantization with extreme compression methods where the approximations introduced by STE are severe, such as Product Quantization. Our proposal is to only quantize a different random subset of weights during each forward, allowing for un…
Qualcomm AI Research, Qualcomm Technologies, Inc.（高通技术公司高通人工智能研究院）；University of California Irvine（加州大学欧文分校）；Qualcomm AI Research, Qualcomm Technologies Netherlands B.V.（高通技术荷兰有限公司高通人工智能研究院）
We present PLONQ, a progressive neural image compression scheme which pushes the boundary of variable bitrate compression by allowing quality scalable coding with a single bitstream. In contrast to existing learned variable bitrate solutions which produce separate bitstreams for each quality, it enables easier rate-control and requires less storage. Leveraging the latent scaling based variable bitrate solution, we introduce nested quantization, a method that defines multiple quantization levels with nested quantization grids, and progressively refines all latents from the coarsest to the fines…
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…
Generalized Radius and Integrated Codebook Transforms for Differentiable Vector Quantization
Columbia University（哥伦比亚大学）；South China Normal University（南方科技大学）；University of Warwick（沃里克大学）；The City College of New York（纽约城市学院）；Hebei Institute of Communications（河北通信学院）
Compressing neural network architectures is important to allow the deployment of models to embedded or mobile devices, and pruning and quantization are the major approaches to compress neural networks nowadays. Both methods benefit when compression parameters are selected specifically for each layer. Finding good combinations of compression parameters, so-called compression policies, is hard as the problem spans an exponentially large search space. Effective compression policies consider the influence of the specific hardware architecture on the used compression methods. We propose an algorith…
IST Austria（奥地利科学技术研究所）；ETH Zurich（苏黎世联邦理工学院）；Neural Magic（神经魔法公司）
We consider the problem of model compression for deep neural networks (DNNs) in the challenging one-shot/post-training setting, in which we are given an accurate trained model, and must compress it without any retraining, based only on a small amount of calibration input data. This problem has become popular in view of the emerging software and hardware support for executing models compressed via pruning and/or quantization with speedup, and well-performing solutions have been proposed independently for both compression approaches. In this paper, we introduce a new compression framework which …
Soft Quantization: Model Compression Via Weight Coupling
Center for the Physics of Biological Function, Princeton, NJ, USA（生物功能物理中心）；Lewis-Sigler Institute, Princeton, NJ, USA（利斯-西格勒研究所）；Joseph Henry Laboratories of Physics, Princeton, NJ, USA（物理约瑟夫·亨利实验室）；Initiative for the Theoretical Sciences, CUNY Graduate Center, New York City, NY, USA（理论科学倡议，CUNY研究生中心）
In this paper, we investigate lossy compression of deep neural networks (DNNs) by weight quantization and lossless source coding for memory-efficient deployment. Whereas the previous work addressed non-universal scalar quantization and entropy coding of DNN weights, we for the first time introduce universal DNN compression by universal vector quantization and universal source coding. In particular, we examine universal randomized lattice quantization of DNNs, which randomizes DNN weights by uniform random dithering before lattice quantization and can perform near-optimally on any source withou…
We consider the problem of deep neural net compression by quantization: given a large, reference net, we want to quantize its real-valued weights using a codebook with $K$ entries so that the training loss of the quantized net is minimal. The codebook can be optimally learned jointly with the net, or fixed, as for binarization or ternarization approaches. Previous work has quantized the weights of the reference net, or incorporated rounding operations in the backpropagation algorithm, but this has no guarantee of converging to a loss-optimal, quantized net. We describe a new approach based on …
