Recurrent neural networks (RNN) are the backbone of many text and speech applications. These architectures are typically made up of several computationally complex components such as; non-linear activation functions, normalization, bi-directional dependence and attention. In order to maintain good accuracy, these components are frequently run using full-precision floating-point computation, making them slow, inefficient and difficult to deploy on edge devices. In addition, the complex nature of these operations makes them challenging to quantize using standard quantization methods without a significant performance drop. We present a quantization-aware training method for obtaining a highly accurate integer-only recurrent neural network (iRNN). Our approach supports layer normalization, attention, and an adaptive piecewise linear (PWL) approximation of activation functions, to serve a wide range of state-of-the-art RNNs. The proposed method enables RNN-based language models to run on edge devices with $2\times$ improvement in runtime, and $4\times$ reduction in model size while maintaining similar accuracy as its full-precision counterpart.

We introduce a method to train Quantized Neural Networks (QNNs) -neural networks with extremely low precision (e.g., 1-bit) weights and activations, at run-time. At traintime the quantized weights and activations are used for computing the parameter gradients. During the forward pass, QNNs drastically reduce memory size and accesses, and replace most arithmetic operations with bit-wise operations. As a result, power consumption is expected to be drastically reduced. We trained QNNs over the MNIST, CIFAR-10, SVHN and ImageNet datasets. The resulting QNNs achieve prediction accuracy comparable to their 32-bit counterparts. For example, our quantized version of AlexNet with 1-bit weights and 2-bit activations achieves 51% top-1 accuracy. Moreover, we quantize the parameter gradients to 6-bits as well which enables gradients computation using only bit-wise operation. Quantized recurrent neural networks were tested over the Penn Treebank dataset, and achieved comparable accuracy as their 32-bit counterparts using only 4-bits. Last but not least, we programmed a binary matrix multiplication GPU kernel with which it is possible to run our MNIST QNN 7 times faster than with an unoptimized GPU kernel, without suffering any loss in classification accuracy. The QNN code is available online.

我们报告了激进的量化策略，这些策略极大地加速了复发性神经网络传感器（RNN-T）的推理。我们使用4位整数表示进行权重和激活，并应用量化意识训练（QAT）来重新训练完整模型（声学编码器和语言模型）并实现近乎ISO的准确性。我们表明，根据网络本地属性量身定制的自定义量化方案对于在限制QAT的计算开销的同时，至关重要。密度比语言模型融合已显示出在RNN-T工作负载上的准确性提高，但严重增加了推理的计算成本。我们表明，我们的量化策略可以使用大型宽度宽度进行假设搜索，同时实现与流媒体兼容的运行时间，并且与完整的Precision模型相比，我们可以实现与流相兼容的运行时间和7.6 $ \ times $的完整模型压缩比。通过硬件仿真，我们估计端到端量化的RNN-T（包括LM Fusion）的3.4 $ \ times $从fp16到INT4，导致实时因子（RTF）为0.06。在NIST HUB5 2000，HUB5 2001和RT-03测试集中，我们保留了与LM Fusion相关的大部分收益，将平均WER提高了$ 1.5％。

由于神经网络变得更加强大，因此在现实世界中部署它们的愿望是一个上升的愿望;然而，神经网络的功率和准确性主要是由于它们的深度和复杂性，使得它们难以部署，尤其是在资源受限的设备中。最近出现了神经网络量化，以满足这种需求通过降低网络的精度来降低神经网络的大小和复杂性。具有较小和更简单的网络，可以在目标硬件的约束中运行神经网络。本文调查了在过去十年中开发的许多神经网络量化技术。基于该调查和神经网络量化技术的比较，我们提出了该地区的未来研究方向。

The rising popularity of intelligent mobile devices and the daunting computational cost of deep learning-based models call for efficient and accurate on-device inference schemes. We propose a quantization scheme that allows inference to be carried out using integer-only arithmetic, which can be implemented more efficiently than floating point inference on commonly available integer-only hardware. We also co-design a training procedure to preserve end-to-end model accuracy post quantization. As a result, the proposed quantization scheme improves the tradeoff between accuracy and on-device latency. The improvements are significant even on MobileNets, a model family known for run-time efficiency, and are demonstrated in ImageNet classification and COCO detection on popular CPUs.

深入学习模型的压缩在将这些模型部署到边缘设备方面具有根本重要性。在压缩期间，在压缩期间结合硬件模型和应用限制可以最大限度地提高优势，但使其专为一种情况而设计。因此，压缩需要自动化。搜索最佳压缩方法参数被认为是一个优化问题。本文介绍了一种多目标硬件感知量化（MohaQ）方法，其将硬件效率和推理误差视为混合精度量化的目标。该方法通过依赖于两个步骤，在很大的搜索空间中评估候选解决方案。首先，应用训练后量化以进行快速解决方案评估。其次，我们提出了一个名为“基于信标的搜索”的搜索技术，仅在搜索空间中重新选出所选解决方案，并将其用作信标以了解刷新对其他解决方案的影响。为了评估优化潜力，我们使用Timit DataSet选择语音识别模型。该模型基于简单的复发单元（SRU），由于其相当大的加速在其他复发单元上。我们应用了我们在两个平台上运行的方法：SILAGO和BETFUSION。实验评估表明，SRU通过训练后量化可以压缩高达8倍，而误差的任何显着增加，误差只有1.5个百分点增加。在Silago上，唯一的搜索发现解决方案分别实现了最大可能加速和节能的80 \％和64 \％，错误的误差增加了0.5个百分点。在BETFUSION上，对于小SRAM尺寸的约束，基于信标的搜索将推断搜索的错误增益减少4个百分点，并且与BitFusion基线相比，可能的达到的加速度增加到47倍。

胶囊网络（CAPSNET）是图像处理的新兴趋势。与卷积神经网络相反，CAPSNET不容易受到对象变形的影响，因为对象的相对空间信息在整个网络中保存。但是，它们的复杂性主要与胶囊结构和动态路由机制有关，这使得以其原始形式部署封闭式以由小型微控制器（MCU）供电的设备几乎是不合理的。在一个智力从云到边缘迅速转移的时代，这种高复杂性对在边缘的采用capsnets的采用构成了严重的挑战。为了解决此问题，我们提出了一个API，用于执行ARM Cortex-M和RISC-V MCUS中的量化capsnet。我们的软件内核扩展了ARM CMSIS-NN和RISC-V PULP-NN，以用8位整数作为操作数支持胶囊操作。随之而来的是，我们提出了一个框架，以执行CAPSNET的训练后量化。结果显示，记忆足迹的减少近75％，准确性损失范围从0.07％到0.18％。在吞吐量方面，我们的ARM Cortex-M API可以分别在仅119.94和90.60毫秒（MS）的中型胶囊和胶囊层执行（STM32H7555ZIT6U，Cortex-M7 @ 480 MHz）。对于GAP-8 SOC（RISC-V RV32IMCXPULP @ 170 MHz），延迟分别降至7.02和38.03 ms。

与变压器架构相关的自我监督学习的最新进展使自然语言处理（NLP）表现出极低的困惑。如此强大的模型需要越来越多的模型大小，因此需要大量的计算和内存足迹。在本文中，我们为大规模生成语言模型提出了一个有效的推理框架。作为减少模型大小的关键，我们通过不均匀的量化方法量化权重。然后，我们提出的称为NUQMM的量化矩阵乘法加速了，该内核可以在压缩比和准确性之间进行广泛的权衡。我们提出的NUQMM不仅减少了每个GPU的延迟，还减少了大LMS的全部推断，因为高压缩比（通过低位量化）减轻了最小所需的GPU数量。我们证明NUQMM可以将GPT-3（175b）模型的推理速度加速约14.4倍，并将能源消耗降低93％。

在本文中，我们提供了一种系统的方法来评估和比较数字信号处理中神经网络层的计算复杂性。我们提供并链接四个软件到硬件的复杂性度量，定义了不同的复杂度指标与层的超参数的关系。本文解释了如何计算这四个指标以进行馈送和经常性层，并定义在这种情况下，我们应该根据我们是否表征了面向更软件或硬件的应用程序来使用特定的度量。新引入的四个指标之一，称为“添加和位移位数（NAB）”，用于异质量化。 NABS不仅表征了操作中使用的位宽的影响，还表征了算术操作中使用的量化类型。我们打算这项工作作为与神经网络在实时数字信号处理中应用相关的复杂性估计级别（目的）的基线，旨在统一计算复杂性估计。

While machine learning is traditionally a resource intensive task, embedded systems, autonomous navigation, and the vision of the Internet of Things fuel the interest in resource-efficient approaches. These approaches aim for a carefully chosen trade-off between performance and resource consumption in terms of computation and energy. The development of such approaches is among the major challenges in current machine learning research and key to ensure a smooth transition of machine learning technology from a scientific environment with virtually unlimited computing resources into everyday's applications. In this article, we provide an overview of the current state of the art of machine learning techniques facilitating these real-world requirements. In particular, we focus on deep neural networks (DNNs), the predominant machine learning models of the past decade. We give a comprehensive overview of the vast literature that can be mainly split into three non-mutually exclusive categories: (i) quantized neural networks, (ii) network pruning, and (iii) structural efficiency. These techniques can be applied during training or as post-processing, and they are widely used to reduce the computational demands in terms of memory footprint, inference speed, and energy efficiency. We also briefly discuss different concepts of embedded hardware for DNNs and their compatibility with machine learning techniques as well as potential for energy and latency reduction. We substantiate our discussion with experiments on well-known benchmark datasets using compression techniques (quantization, pruning) for a set of resource-constrained embedded systems, such as CPUs, GPUs and FPGAs. The obtained results highlight the difficulty of finding good trade-offs between resource efficiency and predictive performance.

Increasing the size of a neural network typically improves accuracy but also increases the memory and compute requirements for training the model. We introduce methodology for training deep neural networks using half-precision floating point numbers, without losing model accuracy or having to modify hyperparameters. This nearly halves memory requirements and, on recent GPUs, speeds up arithmetic. Weights, activations, and gradients are stored in IEEE halfprecision format. Since this format has a narrower range than single-precision we propose three techniques for preventing the loss of critical information. Firstly, we recommend maintaining a single-precision copy of weights that accumulates the gradients after each optimizer step (this copy is rounded to half-precision for the forward-and back-propagation). Secondly, we propose loss-scaling to preserve gradient values with small magnitudes. Thirdly, we use half-precision arithmetic that accumulates into single-precision outputs, which are converted to halfprecision before storing to memory. We demonstrate that the proposed methodology works across a wide variety of tasks and modern large scale (exceeding 100 million parameters) model architectures, trained on large datasets.

在本文中，提出了一种新的方法，该方法允许基于神经网络（NN）均衡器的低复杂性发展，以缓解高速相干光学传输系统中的损伤。在这项工作中，我们提供了已应用于馈电和经常性NN设计的各种深层模型压缩方法的全面描述和比较。此外，我们评估了这些策略对每个NN均衡器的性能的影响。考虑量化，重量聚类，修剪和其他用于模型压缩的尖端策略。在这项工作中，我们提出并评估贝叶斯优化辅助压缩，其中选择了压缩的超参数以同时降低复杂性并提高性能。总之，通过使用模拟和实验数据来评估每种压缩方法的复杂性及其性能之间的权衡，以完成分析。通过利用最佳压缩方法，我们表明可以设计基于NN的均衡器，该均衡器比传统的数字背部传播（DBP）均衡器具有更好的性能，并且只有一个步骤。这是通过减少使用加权聚类和修剪算法后在NN均衡器中使用的乘数数量来完成的。此外，我们证明了基于NN的均衡器也可以实现卓越的性能，同时仍然保持与完整的电子色色散补偿块相同的复杂性。我们通过强调开放问题和现有挑战以及未来的研究方向来结束分析。

基于变压器的模型用于实现各种深度学习任务的最新性能。由于基于变压器的模型具有大量参数，因此在下游任务上进行微调是计算密集型和饥饿的能量。此类型号的自动混合精液FP32/FP16微调以前已用于降低计算资源需求。但是，随着低位整数背面传播的最新进展，有可能进一步减少计算和记忆脚印。在这项工作中，我们探索了一种新颖的整数训练方法，该方法使用整数算术来进行正向传播和梯度计算，对基于变压器的模型中的线性，卷积，层和层和嵌入层的梯度计算。此外，我们研究了各种整数位宽度的效果，以找到基于变压器模型的整数微调所需的最小位宽度。我们使用整数层对流行的下游任务进行了微调和VIT模型。我们表明，16位整数模型与浮点基线性能匹配。将位宽度降低到10，我们观察到0.5平均得分下降。最后，将位宽度的进一步降低到8的平均得分下降为1.7分。

It has been witnessed that learned image compression has outperformed conventional image coding techniques and tends to be practical in industrial applications. One of the most critical issues that need to be considered is the non-deterministic calculation, which makes the probability prediction cross-platform inconsistent and frustrates successful decoding. We propose to solve this problem by introducing well-developed post-training quantization and making the model inference integer-arithmetic-only, which is much simpler than presently existing training and fine-tuning based approaches yet still keeps the superior rate-distortion performance of learned image compression. Based on that, we further improve the discretization of the entropy parameters and extend the deterministic inference to fit Gaussian mixture models. With our proposed methods, the current state-of-the-art image compression models can infer in a cross-platform consistent manner, which makes the further development and practice of learned image compression more promising.

Neural Machine Translation (NMT) is an end-to-end learning approach for automated translation, with the potential to overcome many of the weaknesses of conventional phrase-based translation systems. Unfortunately, NMT systems are known to be computationally expensive both in training and in translation inference -sometimes prohibitively so in the case of very large data sets and large models. Several authors have also charged that NMT systems lack robustness, particularly when input sentences contain rare words. These issues have hindered NMT's use in practical deployments and services, where both accuracy and speed are essential. In this work, we present GNMT, Google's Neural Machine Translation system, which attempts to address many of these issues. Our model consists of a deep LSTM network with 8 encoder and 8 decoder layers using residual connections as well as attention connections from the decoder network to the encoder. To improve parallelism and therefore decrease training time, our attention mechanism connects the bottom layer of the decoder to the top layer of the encoder. To accelerate the final translation speed, we employ low-precision arithmetic during inference computations. To improve handling of rare words, we divide words into a limited set of common sub-word units ("wordpieces") for both input and output. This method provides a good balance between the flexibility of "character"-delimited models and the efficiency of "word"-delimited models, naturally handles translation of rare words, and ultimately improves the overall accuracy of the system. Our beam search technique employs a length-normalization procedure and uses a coverage penalty, which encourages generation of an output sentence that is most likely to cover all the words in the source sentence. To directly optimize the translation BLEU scores, we consider refining the models by using reinforcement learning, but we found that the improvement in the BLEU scores did not reflect in the human evaluation. On the WMT'14 English-to-French and English-to-German benchmarks, GNMT achieves competitive results to state-of-the-art. Using a human side-by-side evaluation on a set of isolated simple sentences, it reduces translation errors by an average of 60% compared to Google's phrase-based production system.

代表低精度的深度神经网络（DNN）是一种有希望的方法来实现有效的加速和记忆力。以前的方法在低精度中培训DNN的方法通常在重量更新期间在高精度中保持重量的重量副本。由于低精度数字系统与学习算法之间的复杂相互作用，直接具有低精度重量的培训导致精度下降。为了解决这个问题，我们开发了一个共同设计的低精度训练框架，被称为LNS-MADAM，我们共同设计了对数号系统（LNS）和乘法权重算法（MADAM）。我们证明了LNS-MADAM在重量更新期间导致低量化误差，即使精度有限，也导致稳定的收敛。我们进一步提出了LNS-MADAM的硬件设计，可以解决实现LNS计算的有效数据路径的实际挑战。我们的实现有效地降低了LNS - 整数转换和部分总和累积所产生的能量开销。实验结果表明，LNS-MADAM为全精密对应物达到了可比的准确性，只有8位对流行的计算机视觉和自然语言任务。与全精密浮点实施相比，LNS-MADAM将能耗降低超过90。

混合精确的深神经网络达到了硬件部署所需的能源效率和吞吐量，尤其是在资源有限的情况下，而无需牺牲准确性。但是，不容易找到保留精度的最佳每层钻头精度，尤其是在创建巨大搜索空间的大量模型，数据集和量化技术中。为了解决这一困难，最近出现了一系列文献，并且已经提出了一些实现有希望的准确性结果的框架。在本文中，我们首先总结了文献中通常使用的量化技术。然后，我们对混合精液框架进行了彻底的调查，该调查是根据其优化技术进行分类的，例如增强学习和量化技术，例如确定性舍入。此外，讨论了每个框架的优势和缺点，我们在其中呈现并列。我们最终为未来的混合精液框架提供了指南。

深度学习模型的计算复杂性不断增加，使他们在各种云和边缘平台上的培训和部署变得困难。用低位整数算术代替浮点算术是一种有希望的方法，可节省能量，记忆足迹和深度学习模型的延迟。因此，量化引起了近年来研究人员的注意。但是，没有详细研究使用整数数字形成功能齐全的整数训练管道，包括前进，后传播和随机梯度下降。我们的经验和数学结果表明，整数算术足以训练深度学习模型。与最近的建议不同，我们直接切换计算的数字表示。我们的新型训练方法形成了完全整数训练管道，与浮点相比，它不会改变损失和准确性的轨迹，也不需要任何特殊的超参数调整，分配调整或梯度剪辑。我们的实验结果表明，我们提出的方法在各种任务（包括视觉变压器），对象检测和语义分割等多种任务中有效。

The pre-dominant approach to language modeling to date is based on recurrent neural networks. Their success on this task is often linked to their ability to capture unbounded context. In this paper we develop a finite context approach through stacked convolutions, which can be more efficient since they allow parallelization over sequential tokens. We propose a novel simplified gating mechanism that outperforms Oord et al. (2016b) and investigate the impact of key architectural decisions. The proposed approach achieves state-of-the-art on the WikiText-103 benchmark, even though it features longterm dependencies, as well as competitive results on the Google Billion Words benchmark. Our model reduces the latency to score a sentence by an order of magnitude compared to a recurrent baseline. To our knowledge, this is the first time a non-recurrent approach is competitive with strong recurrent models on these large scale language tasks.

模型量化已成为加速深度学习推理的不可或缺的技术。虽然研究人员继续推动量化算法的前沿，但是现有量化工作通常是不可否认的和不可推销的。这是因为研究人员不选择一致的训练管道并忽略硬件部署的要求。在这项工作中，我们提出了模型量化基准（MQBench），首次尝试评估，分析和基准模型量化算法的再现性和部署性。我们为实际部署选择多个不同的平台，包括CPU，GPU，ASIC，DSP，并在统一培训管道下评估广泛的最新量化算法。 MQBENCK就像一个连接算法和硬件的桥梁。我们进行全面的分析，并找到相当大的直观或反向直观的见解。通过对齐训练设置，我们发现现有的算法在传统的学术轨道上具有大致相同的性能。虽然用于硬件可部署量化，但有一个巨大的精度差距，仍然不稳定。令人惊讶的是，没有现有的算法在MQBench中赢得每一项挑战，我们希望这项工作能够激发未来的研究方向。