Transfer learning is a popular technique for improving the performance of neural networks. However, existing methods are limited to transferring parameters between networks with same architectures. We present a method for transferring parameters between neural networks with different architectures. Our method, called DPIAT, uses dynamic programming to match blocks and layers between architectures and transfer parameters efficiently. Compared to existing parameter prediction and random initialization methods, it significantly improves training efficiency and validation accuracy. In experiments on ImageNet, our method improved validation accuracy by an average of 1.6 times after 50 epochs of training. DPIAT allows both researchers and neural architecture search systems to modify trained networks and reuse knowledge, avoiding the need for retraining from scratch. We also introduce a network architecture similarity measure, enabling users to choose the best source network without any training.

The automated machine learning (AutoML) field has become increasingly relevant in recent years. These algorithms can develop models without the need for expert knowledge, facilitating the application of machine learning techniques in the industry. Neural Architecture Search (NAS) exploits deep learning techniques to autonomously produce neural network architectures whose results rival the state-of-the-art models hand-crafted by AI experts. However, this approach requires significant computational resources and hardware investments, making it less appealing for real-usage applications. This article presents the third version of Pareto-Optimal Progressive Neural Architecture Search (POPNASv3), a new sequential model-based optimization NAS algorithm targeting different hardware environments and multiple classification tasks. Our method is able to find competitive architectures within large search spaces, while keeping a flexible structure and data processing pipeline to adapt to different tasks. The algorithm employs Pareto optimality to reduce the number of architectures sampled during the search, drastically improving the time efficiency without loss in accuracy. The experiments performed on images and time series classification datasets provide evidence that POPNASv3 can explore a large set of assorted operators and converge to optimal architectures suited for the type of data provided under different scenarios.

深度学习技术在各种任务中都表现出了出色的有效性，并且深度学习具有推进多种应用程序（包括在边缘计算中）的潜力，其中将深层模型部署在边缘设备上，以实现即时的数据处理和响应。一个关键的挑战是，虽然深层模型的应用通常会产生大量的内存和计算成本，但Edge设备通常只提供非常有限的存储和计算功能，这些功能可能会在各个设备之间差异很大。这些特征使得难以构建深度学习解决方案，以释放边缘设备的潜力，同时遵守其约束。应对这一挑战的一种有希望的方法是自动化有效的深度学习模型的设计，这些模型轻巧，仅需少量存储，并且仅产生低计算开销。该调查提供了针对边缘计算的深度学习模型设计自动化技术的全面覆盖。它提供了关键指标的概述和比较，这些指标通常用于量化模型在有效性，轻度和计算成本方面的水平。然后，该调查涵盖了深层设计自动化技术的三类最新技术：自动化神经体系结构搜索，自动化模型压缩以及联合自动化设计和压缩。最后，调查涵盖了未来研究的开放问题和方向。

传输学习方法旨在使用在丰富的源域上掠过的模型来提高数据稀缺目标域中的性能。一种成本效益的策略，线性探测涉及冻结源模型并培训目标域的新分类头。此策略的表现优于更昂贵但最先进的方法 - 将源模型的所有参数微调到目标域 - 可能是因为微调允许模型从中间层利用有用的信息否则被稍后的净化层丢弃。我们探讨了这些中间层可能直接剥削的假设。我们提出了一种方法，头对脚趾探测（Head2ToE），其从源模型的所有层中选择特征，以训练目标域的分类头。在VTAB-1K的评估中，Head2Toe与平均微调获得的性能相匹配，同时减少培训和储存成本一百倍或更多，但批判性地，用于分配转移，头部2ToE优于微调。

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。

在实践中，非常苛刻，有时无法收集足够大的标记数据数据集以成功培训机器学习模型，并且对此问题的一个可能解决方案是转移学习。本研究旨在评估如何可转让的时间序列数据和哪些条件下的不同域之间的特征。在训练期间，在模型的预测性能和收敛速度方面观察到转移学习的影响。在我们的实验中，我们使用1,500和9,000个数据实例的减少数据集来模仿现实世界的条件。使用相同的缩小数据集，我们培训了两组机器学习模型：那些随着转移学习的培训和从头开始培训的机器学习模型。使用四台机器学习模型进行实验。在相同的应用领域（地震学）以及相互不同的应用领域（地震，语音，医学，金融）之间进行知识转移。我们在训练期间遵守模型的预测性能和收敛速度。为了确认所获得的结果的有效性，我们重复了实验七次并应用了统计测试以确认结果的重要性。我们研究的一般性结论是转移学习可能会增加或不会对模型的预测性能或其收敛速度产生负面影响。在更多细节中分析收集的数据，以确定哪些源域和目标域兼容以用于传输知识。我们还分析了目标数据集大小的效果和模型的选择及其超参数对转移学习的影响。

In the past few years, neural architecture search (NAS) has become an increasingly important tool within the deep learning community. Despite the many recent successes of NAS, however, most existing approaches operate within highly structured design spaces, and hence explore only a small fraction of the full search space of neural architectures while also requiring significant manual effort from domain experts. In this work, we develop techniques that enable efficient NAS in a significantly larger design space. To accomplish this, we propose to perform NAS in an abstract search space of program properties. Our key insights are as follows: (1) the abstract search space is significantly smaller than the original search space, and (2) architectures with similar program properties also have similar performance; thus, we can search more efficiently in the abstract search space. To enable this approach, we also propose a novel efficient synthesis procedure, which accepts a set of promising program properties, and returns a satisfying neural architecture. We implement our approach, $\alpha$NAS, within an evolutionary framework, where the mutations are guided by the program properties. Starting with a ResNet-34 model, $\alpha$NAS produces a model with slightly improved accuracy on CIFAR-10 but 96% fewer parameters. On ImageNet, $\alpha$NAS is able to improve over Vision Transformer (30% fewer FLOPS and parameters), ResNet-50 (23% fewer FLOPS, 14% fewer parameters), and EfficientNet (7% fewer FLOPS and parameters) without any degradation in accuracy.

使用DataSet的真实标签培训而不是随机标签导致更快的优化和更好的泛化。这种差异归因于自然数据集中的输入和标签之间的对齐概念。我们发现，随机或真正标签上的具有不同架构和优化器的培训神经网络在隐藏的表示和训练标签之间强制执行相同的关系，阐明为什么神经网络表示为转移如此成功。我们首先突出显示为什么对齐的特征在经典的合成转移问题中促进转移和展示，即对齐是对相似和不同意任务的正负传输的确定因素。然后我们调查各种神经网络架构，并发现（a）在各种不同的架构和优化器中出现的对齐，并且从深度（b）对准产生的更多对准对于更接近输出的层和（c）现有的性能深度CNN表现出高级别的对准。

我们介绍了延迟感知网络加速度（LANA） - 一种在神经结构上建立的方法，用于加速神经网络的神经结构搜索技术和教师学生蒸馏。 Lana由两个阶段组成：在第一阶段，它会使用层面特征映射蒸馏来列举每层教师网络的许多替代操作。在第二阶段，它解决了使用新颖的整数线性优化（ILP）方法的有效操作的组合选择。 ILP带来独特的属性，因为它（i）在几秒钟内执行NAS，（ii）轻松满足预算约束，（iii）在图层粒度上工作，（iv）支持巨大的搜索空间$ o（10 ^ { 100}）$，超越先前的搜索方法，效率和效率。在广泛的实验中，我们表明Lana产生了由目标潜伏期预算限制的有效和准确的模型，同时比其他技术明显快。我们分析了三个流行的网络架构：高效的网络，高效网络和reses，并在压缩较大模型的较小模型的延迟级别时，实现所有型号（高达3.0 \％$）的准确性改进。 Lana通过GPU和CPU实现显着的加速（高达5美元\倍），以没有准确性下降。代码将很快分享。

深度学习领域的最新进展表明，非常大的神经网络在几种应用中的有效性。但是，随着这些深度神经网络的大小不断增长，配置其许多参数以获得良好的结果变得越来越困难。目前，分析师必须尝试许多不同的配置和参数设置，这些配置和参数设置是劳动密集型且耗时的。另一方面，没有人类专家的领域知识，用于神经网络架构搜索的完全自动化技术的能力受到限制。为了解决问题，我们根据单次体系结构搜索技术制定神经网络体系结构优化的任务作为图形空间探索。在这种方法中，对所有候选体系结构的超级绘制进行了一次训练，并将最佳神经网络确定为子图。在本文中，我们提出了一个框架，该框架允许分析师有效地构建解决方案子图形空间，并通过注入其域知识来指导网络搜索。从由基本神经网络组件组成的网络体系结构空间开始，分析师有权通过我们的单发搜索方案有效地选择最有希望的组件。以迭代方式应用此技术使分析师可以为给定应用程序收敛到最佳性能的神经网络体系结构。在探索过程中，分析师可以利用其域知识在搜索空间的散点图可视化中提供的线索来帮助编辑不同的组件，并指导搜索更快的融合。我们与几位深度学习研究人员合作设计了界面，并通过用户研究和两个案例研究来评估其最终有效性。

Multitask learning assumes that models capable of learning from multiple tasks can achieve better quality and efficiency via knowledge transfer, a key feature of human learning. Though, state of the art ML models rely on high customization for each task and leverage size and data scale rather than scaling the number of tasks. Also, continual learning, that adds the temporal aspect to multitask, is often focused to the study of common pitfalls such as catastrophic forgetting instead of being studied at a large scale as a critical component to build the next generation artificial intelligence.We propose an evolutionary method capable of generating large scale multitask models that support the dynamic addition of new tasks. The generated multitask models are sparsely activated and integrates a task-based routing that guarantees bounded compute cost and fewer added parameters per task as the model expands.The proposed method relies on a knowledge compartmentalization technique to achieve immunity against catastrophic forgetting and other common pitfalls such as gradient interference and negative transfer. We demonstrate empirically that the proposed method can jointly solve and achieve competitive results on 69public image classification tasks, for example improving the state of the art on a competitive benchmark such as cifar10 by achieving a 15% relative error reduction compared to the best model trained on public data.

Neural networks require careful weight initialization to prevent signals from exploding or vanishing. Existing initialization schemes solve this problem in specific cases by assuming that the network has a certain activation function or topology. It is difficult to derive such weight initialization strategies, and modern architectures therefore often use these same initialization schemes even though their assumptions do not hold. This paper introduces AutoInit, a weight initialization algorithm that automatically adapts to different neural network architectures. By analytically tracking the mean and variance of signals as they propagate through the network, AutoInit appropriately scales the weights at each layer to avoid exploding or vanishing signals. Experiments demonstrate that AutoInit improves performance of convolutional, residual, and transformer networks across a range of activation function, dropout, weight decay, learning rate, and normalizer settings, and does so more reliably than data-dependent initialization methods. This flexibility allows AutoInit to initialize models for everything from small tabular tasks to large datasets such as ImageNet. Such generality turns out particularly useful in neural architecture search and in activation function discovery. In these settings, AutoInit initializes each candidate appropriately, making performance evaluations more accurate. AutoInit thus serves as an automatic configuration tool that makes design of new neural network architectures more robust. The AutoInit package provides a wrapper around TensorFlow models and is available at https://github.com/cognizant-ai-labs/autoinit.

Alphazero，Leela Chess Zero和Stockfish Nnue革新了计算机国际象棋。本书对此类引擎的技术内部工作进行了完整的介绍。该书分为四个主要章节 - 不包括第1章（简介）和第6章（结论）：第2章引入神经网络，涵盖了所有用于构建深层网络的基本构建块，例如Alphazero使用的网络。内容包括感知器，后传播和梯度下降，分类，回归，多层感知器，矢量化技术，卷积网络，挤压网络，挤压和激发网络，完全连接的网络，批处理归一化和横向归一化和跨性线性单位，残留层，剩余层，过度效果和底漆。第3章介绍了用于国际象棋发动机以及Alphazero使用的经典搜索技术。内容包括minimax，alpha-beta搜索和蒙特卡洛树搜索。第4章展示了现代国际象棋发动机的设计。除了开创性的Alphago，Alphago Zero和Alphazero我们涵盖Leela Chess Zero，Fat Fritz，Fat Fritz 2以及有效更新的神经网络（NNUE）以及MAIA。第5章是关于实施微型α。 Shexapawn是国际象棋的简约版本，被用作为此的示例。 Minimax搜索可以解决六ap峰，并产生了监督学习的培训位置。然后，作为比较，实施了类似Alphazero的训练回路，其中通过自我游戏进行训练与强化学习结合在一起。最后，比较了类似α的培训和监督培训。

Deep residual networks were shown to be able to scale up to thousands of layers and still have improving performance. However, each fraction of a percent of improved accuracy costs nearly doubling the number of layers, and so training very deep residual networks has a problem of diminishing feature reuse, which makes these networks very slow to train. To tackle these problems, in this paper we conduct a detailed experimental study on the architecture of ResNet blocks, based on which we propose a novel architecture where we decrease depth and increase width of residual networks. We call the resulting network structures wide residual networks (WRNs) and show that these are far superior over their commonly used thin and very deep counterparts. For example, we demonstrate that even a simple 16-layer-deep wide residual network outperforms in accuracy and efficiency all previous deep residual networks, including thousand-layerdeep networks, achieving new state-of-the-art results on CIFAR, SVHN, COCO, and significant improvements on ImageNet. Our code and models are available at https: //github.com/szagoruyko/wide-residual-networks.

We propose a new method for learning the structure of convolutional neural networks (CNNs) that is more efficient than recent state-of-the-art methods based on reinforcement learning and evolutionary algorithms. Our approach uses a sequential model-based optimization (SMBO) strategy, in which we search for structures in order of increasing complexity, while simultaneously learning a surrogate model to guide the search through structure space. Direct comparison under the same search space shows that our method is up to 5 times more efficient than the RL method of Zoph et al. (2018) in terms of number of models evaluated, and 8 times faster in terms of total compute. The structures we discover in this way achieve state of the art classification accuracies on CIFAR-10 and ImageNet.

Does the dominant approach to learn representations (as a side effect of optimizing an expected cost for a single training distribution) remain a good approach when we are dealing with multiple distributions. Our thesis is that such scenarios are better served by representations that are "richer" than those obtained with a single optimization episode. This is supported by a collection of empirical results obtained with an apparently na\"ive ensembling technique: concatenating the representations obtained with multiple training episodes using the same data, model, algorithm, and hyper-parameters, but different random seeds. These independently trained networks perform similarly. Yet, in a number of scenarios involving new distributions, the concatenated representation performs substantially better than an equivalently sized network trained from scratch. This proves that the representations constructed by multiple training episodes are in fact different. Although their concatenation carries little additional information about the training task under the training distribution, it becomes substantially more informative when tasks or distributions change. Meanwhile, a single training episode is unlikely to yield such a redundant representation because the optimization process has no reason to accumulate features that do not incrementally improve the training performance.

以前的工作提出了许多新的损失函数和常规程序，可提高图像分类任务的测试准确性。但是，目前尚不清楚这些损失函数是否了解下游任务的更好表示。本文研究了培训目标的选择如何影响卷积神经网络隐藏表示的可转移性，训练在想象中。我们展示了许多目标在Vanilla Softmax交叉熵上导致想象的精度有统计学意义的改进，但由此产生的固定特征提取器转移到下游任务基本较差，并且当网络完全微调时，损失的选择几乎没有效果新任务。使用居中内核对齐来测量网络隐藏表示之间的相似性，我们发现损失函数之间的差异仅在网络的最后几层中都很明显。我们深入了解倒数第二层的陈述，发现不同的目标和近奇计的组合导致大幅不同的类别分离。具有较高类别分离的表示可以在原始任务上获得更高的准确性，但它们的功能对于下游任务不太有用。我们的结果表明，用于原始任务的学习不变功能与传输任务相关的功能之间存在权衡。

Pruning refers to the elimination of trivial weights from neural networks. The sub-networks within an overparameterized model produced after pruning are often called Lottery tickets. This research aims to generate winning lottery tickets from a set of lottery tickets that can achieve similar accuracy to the original unpruned network. We introduce a novel winning ticket called Cyclic Overlapping Lottery Ticket (COLT) by data splitting and cyclic retraining of the pruned network from scratch. We apply a cyclic pruning algorithm that keeps only the overlapping weights of different pruned models trained on different data segments. Our results demonstrate that COLT can achieve similar accuracies (obtained by the unpruned model) while maintaining high sparsities. We show that the accuracy of COLT is on par with the winning tickets of Lottery Ticket Hypothesis (LTH) and, at times, is better. Moreover, COLTs can be generated using fewer iterations than tickets generated by the popular Iterative Magnitude Pruning (IMP) method. In addition, we also notice COLTs generated on large datasets can be transferred to small ones without compromising performance, demonstrating its generalizing capability. We conduct all our experiments on Cifar-10, Cifar-100 & TinyImageNet datasets and report superior performance than the state-of-the-art methods.

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.

Vision Transformers convert images to sequences by slicing them into patches. The size of these patches controls a speed/accuracy tradeoff, with smaller patches leading to higher accuracy at greater computational cost, but changing the patch size typically requires retraining the model. In this paper, we demonstrate that simply randomizing the patch size at training time leads to a single set of weights that performs well across a wide range of patch sizes, making it possible to tailor the model to different compute budgets at deployment time. We extensively evaluate the resulting model, which we call FlexiViT, on a wide range of tasks, including classification, image-text retrieval, open-world detection, panoptic segmentation, and semantic segmentation, concluding that it usually matches, and sometimes outperforms, standard ViT models trained at a single patch size in an otherwise identical setup. Hence, FlexiViT training is a simple drop-in improvement for ViT that makes it easy to add compute-adaptive capabilities to most models relying on a ViT backbone architecture. Code and pre-trained models are available at https://github.com/google-research/big_vision