Using graph neural networks for large graphs is challenging since there is no clear way of constructing mini-batches. To solve this, previous methods have relied on sampling or graph clustering. While these approaches often lead to good training convergence, they introduce significant overhead due to expensive random data accesses and perform poorly during inference. In this work we instead focus on model behavior during inference. We theoretically model batch construction via maximizing the influence score of nodes on the outputs. This formulation leads to optimal approximation of the output when we do not have knowledge of the trained model. We call the resulting method influence-based mini-batching (IBMB). IBMB accelerates inference by up to 130x compared to previous methods that reach similar accuracy. Remarkably, with adaptive optimization and the right training schedule IBMB can also substantially accelerate training, thanks to precomputed batches and consecutive memory accesses. This results in up to 18x faster training per epoch and up to 17x faster convergence per runtime compared to previous methods.
translated by 谷歌翻译
Neural message passing algorithms for semi-supervised classification on graphs have recently achieved great success. However, for classifying a node these methods only consider nodes that are a few propagation steps away and the size of this utilized neighborhood is hard to extend. In this paper, we use the relationship between graph convolutional networks (GCN) and PageRank to derive an improved propagation scheme based on personalized PageRank. We utilize this propagation procedure to construct a simple model, personalized propagation of neural predictions (PPNP), and its fast approximation, APPNP. Our model's training time is on par or faster and its number of parameters on par or lower than previous models. It leverages a large, adjustable neighborhood for classification and can be easily combined with any neural network. We show that this model outperforms several recently proposed methods for semi-supervised classification in the most thorough study done so far for GCN-like models. Our implementation is available online. 1
translated by 谷歌翻译
Graph convolutional network (GCN) has been successfully applied to many graph-based applications; however, training a large-scale GCN remains challenging. Current SGD-based algorithms suffer from either a high computational cost that exponentially grows with number of GCN layers, or a large space requirement for keeping the entire graph and the embedding of each node in memory. In this paper, we propose Cluster-GCN, a novel GCN algorithm that is suitable for SGD-based training by exploiting the graph clustering structure. Cluster-GCN works as the following: at each step, it samples a block of nodes that associate with a dense subgraph identified by a graph clustering algorithm, and restricts the neighborhood search within this subgraph. This simple but effective strategy leads to significantly improved memory and computational efficiency while being able to achieve comparable test accuracy with previous algorithms. To test the scalability of our algorithm, we create a new Amazon2M data with 2 million nodes and 61 million edges which is more than 5 times larger than the previous largest publicly available dataset (Reddit). For training a 3-layer GCN on this data, Cluster-GCN is faster than the previous state-of-the-art VR-GCN (1523 seconds vs 1961 seconds) and using much less memory (2.2GB vs 11.2GB). Furthermore, for training 4 layer GCN on this data, our algorithm can finish in around 36 minutes while all the existing GCN training algorithms fail to train due to the out-of-memory issue. Furthermore, Cluster-GCN allows us to train much deeper GCN without much time and memory overhead, which leads to improved prediction accuracy-using a 5-layer Cluster-GCN, we achieve state-of-the-art test F1 score 99.36 on the PPI dataset, while the previous best result was 98.71 by [16]. Our codes are publicly available at https://github.com/google-research/google-research/ tree/master/cluster_gcn.
translated by 谷歌翻译
鉴于他们的普及和应用程序的多样性,图形神经网络(GNNS)越来越重要。然而,对对抗性袭击的脆弱性的现有研究依赖于相对较小的图形。我们解决了这个差距并研究了如何在规模攻击和捍卫GNN。我们提出了两个稀疏感知的一阶优化攻击,尽管优化了在节点数量中的许多参数上优化了有效的表示。我们表明,普通的替代损失并不适合全球对GNN的攻击。我们的替代品可以加倍攻击力量。此外,为了提高GNNS的可靠性,我们设计了强大的聚合函数,软中位,导致所有尺度的有效防御。我们评估了我们的攻击和防御与图形的标准GNN,与以前的工作相比大于100倍以上。我们甚至通过将技术扩展到可伸缩的GNN来进一步缩放一个数量级。
translated by 谷歌翻译
Graph Convolutional Networks (GCNs) are powerful models for learning representations of attributed graphs. To scale GCNs to large graphs, state-of-the-art methods use various layer sampling techniques to alleviate the "neighbor explosion" problem during minibatch training. We propose GraphSAINT, a graph sampling based inductive learning method that improves training efficiency and accuracy in a fundamentally different way. By changing perspective, GraphSAINT constructs minibatches by sampling the training graph, rather than the nodes or edges across GCN layers. Each iteration, a complete GCN is built from the properly sampled subgraph. Thus, we ensure fixed number of well-connected nodes in all layers. We further propose normalization technique to eliminate bias, and sampling algorithms for variance reduction. Importantly, we can decouple the sampling from the forward and backward propagation, and extend GraphSAINT with many architecture variants (e.g., graph attention, jumping connection). GraphSAINT demonstrates superior performance in both accuracy and training time on five large graphs, and achieves new state-of-the-art F1 scores for PPI (0.995) and Reddit (0.970).
translated by 谷歌翻译
图形神经网络(GNN)已被证明是分析非欧国人图数据的强大工具。但是,缺乏有效的分布图学习(GL)系统极大地阻碍了GNN的应用,尤其是当图形大且GNN相对深时。本文中,我们提出了GraphTheta,这是一种以顶点为中心的图形编程模型实现的新颖分布式和可扩展的GL系统。 GraphTheta是第一个基于分布式图处理的GL系统,其神经网络运算符以用户定义的功能实现。该系统支持多种培训策略,并在分布式(虚拟)机器上启用高度可扩展的大图学习。为了促进图形卷积实现,GraphTheta提出了一个名为NN-Tgar的新的GL抽象,以弥合图形处理和图形深度学习之间的差距。提出了分布式图引擎,以通过混合平行执行进行随机梯度下降优化。此外,除了全球批次和迷你批次外,我们还为新的集群批次培训策略提供了支持。我们使用许多网络大小的数据集评估GraphTheta,范围从小,适度到大规模。实验结果表明,GraphTheta可以很好地扩展到1,024名工人,用于培训内部开发的GNN,该工业尺度的Aripay数据集为14亿个节点和41亿个属性边缘,并带有CPU虚拟机(Dockers)群的小群。 (5 $ \ sim $ 12GB)。此外,GraphTheta比最先进的GNN实现获得了可比或更好的预测结果,证明其学习GNN和现有框架的能力,并且可以超过多达$ 2.02 \ tims $ $ 2.02 \ times $,具有更好的可扩展性。据我们所知,这项工作介绍了文献中最大的边缘属性GNN学习任务。
translated by 谷歌翻译
数据处理的最新进展刺激了对非常大尺度的学习图的需求。众所周知,图形神经网络(GNN)是解决图形学习任务的一种新兴和有力的方法,很难扩大规模。大多数可扩展模型应用基于节点的技术来简化GNN的昂贵图形消息传播过程。但是,我们发现当应用于百万甚至数十亿尺度的图表时,这种加速度不足。在这项工作中,我们提出了Scara,这是一种可扩展的GNN,具有针对图形计算的特征优化。 Scara有效地计算出从节点功能中嵌入的图形,并进一步选择和重用功能计算结果以减少开销。理论分析表明,我们的模型在传播过程以及GNN培训和推理中具有确保精度,实现了子线性时间的复杂性。我们在各种数据集上进行了广泛的实验,以评估圣aca的功效和效率。与基线的性能比较表明,与快速收敛和可比精度相比,与当前的最新方法相比,圣aca最高可达到100倍的图形传播加速度。最值得注意的是,在100秒内处理最大的十亿个GNN数据集纸100m(1.11亿节点,1.6B边缘)上的预先计算是有效的。
translated by 谷歌翻译
图形神经网络(GNNS)将深度神经网络(DNN)的成功扩展到非欧几里德图数据,实现了各种任务的接地性能,例如节点分类和图形属性预测。尽管如此,现有系统效率低,培训数十亿节点和GPU的节点和边缘训练大图。主要瓶颈是准备GPU数据的过程 - 子图采样和特征检索。本文提出了一个分布式GNN培训系统的BGL,旨在解决一些关键思想的瓶颈。首先,我们提出了一种动态缓存引擎,以最小化特征检索流量。通过协同设计缓存政策和抽样顺序,我们发现低开销和高缓存命中率的精美斑点。其次,我们改善了曲线图分区算法,以减少子图采样期间的交叉分区通信。最后,仔细资源隔离减少了不同数据预处理阶段之间的争用。关于各种GNN模型和大图数据集的广泛实验表明,BGL平均明显优于现有的GNN训练系统20.68倍。
translated by 谷歌翻译
图形神经网络(GNN)在许多基于图的应用程序中取得了巨大成功。但是,巨大的尺寸和高稀疏度的图表阻碍了其在工业场景下的应用。尽管为大规模图提出了一些可扩展的GNN,但它们为每个节点采用固定的$ k $ hop邻域,因此在稀疏区域内采用大型繁殖深度时面临过度光滑的问题。为了解决上述问题,我们提出了一种新的GNN体系结构 - 图形注意多层感知器(GAMLP),该架构可以捕获不同图形知识范围之间的基本相关性。我们已经与天使平台部署了GAMLP,并进一步评估了现实世界数据集和大规模工业数据集的GAMLP。这14个图数据集的广泛实验表明,GAMLP在享有高可扩展性和效率的同时,达到了最先进的性能。具体来说,在我们的大规模腾讯视频数据集上的预测准确性方面,它的表现优于1.3 \%,同时达到了高达$ 50 \ times $ triending的速度。此外,它在开放图基准的最大同质和异质图(即OGBN-PAPERS100M和OGBN-MAG)的排行榜上排名第一。
translated by 谷歌翻译
图表表示学习是一种快速增长的领域,其中一个主要目标是在低维空间中产生有意义的图形表示。已经成功地应用了学习的嵌入式来执行各种预测任务,例如链路预测,节点分类,群集和可视化。图表社区的集体努力提供了数百种方法,但在所有评估指标下没有单一方法擅长,例如预测准确性,运行时间,可扩展性等。该调查旨在通过考虑算法来评估嵌入方法的所有主要类别的图表变体,参数选择,可伸缩性,硬件和软件平台,下游ML任务和多样化数据集。我们使用包含手动特征工程,矩阵分解,浅神经网络和深图卷积网络的分类法组织了图形嵌入技术。我们使用广泛使用的基准图表评估了节点分类,链路预测,群集和可视化任务的这些类别算法。我们在Pytorch几何和DGL库上设计了我们的实验,并在不同的多核CPU和GPU平台上运行实验。我们严格地审查了各种性能指标下嵌入方法的性能,并总结了结果。因此,本文可以作为比较指南,以帮助用户选择最适合其任务的方法。
translated by 谷歌翻译
图表卷积网络(GCNS)在各种半监督节点分类任务中取得了令人印象深刻的实证进步。尽管取得了巨大的成功,但在大型图形上培训GCNS遭受了计算和内存问题。规避这些障碍的潜在路径是基于采样的方法,其中在每个层处采样节点的子集。虽然最近的研究已经证明了基于采样的方法的有效性,但这些作品缺乏在现实环境下的理论融合担保,并且不能完全利用优化期间演出参数的信息。在本文中,我们描述并分析了一般的双差异减少模式,可以在内存预算下加速任何采样方法。所提出的模式的激励推动是仔细分析采样方法的差异,其中示出了诱导方差可以在前进传播期间分解为节点嵌入近似方差(Zeroth阶差异)(第一 - 顺序变化)在后向传播期间。理论上,从理论上分析所提出的架构的融合,并显示它享有$ \ Mathcal {O}(1 / T)$收敛率。我们通过将建议的模式集成在不同的采样方法中并将其应用于不同的大型实际图形来补充我们的理论结果。
translated by 谷歌翻译
Graph Convolutional Networks (GCNs) are extensively utilized for deep learning on graphs. The large data sizes of graphs and their vertex features make scalable training algorithms and distributed memory systems necessary. Since the convolution operation on graphs induces irregular memory access patterns, designing a memory- and communication-efficient parallel algorithm for GCN training poses unique challenges. We propose a highly parallel training algorithm that scales to large processor counts. In our solution, the large adjacency and vertex-feature matrices are partitioned among processors. We exploit the vertex-partitioning of the graph to use non-blocking point-to-point communication operations between processors for better scalability. To further minimize the parallelization overheads, we introduce a sparse matrix partitioning scheme based on a hypergraph partitioning model for full-batch training. We also propose a novel stochastic hypergraph model to encode the expected communication volume in mini-batch training. We show the merits of the hypergraph model, previously unexplored for GCN training, over the standard graph partitioning model which does not accurately encode the communication costs. Experiments performed on real-world graph datasets demonstrate that the proposed algorithms achieve considerable speedups over alternative solutions. The optimizations achieved on communication costs become even more pronounced at high scalability with many processors. The performance benefits are preserved in deeper GCNs having more layers as well as on billion-scale graphs.
translated by 谷歌翻译
Graph classification is an important area in both modern research and industry. Multiple applications, especially in chemistry and novel drug discovery, encourage rapid development of machine learning models in this area. To keep up with the pace of new research, proper experimental design, fair evaluation, and independent benchmarks are essential. Design of strong baselines is an indispensable element of such works. In this thesis, we explore multiple approaches to graph classification. We focus on Graph Neural Networks (GNNs), which emerged as a de facto standard deep learning technique for graph representation learning. Classical approaches, such as graph descriptors and molecular fingerprints, are also addressed. We design fair evaluation experimental protocol and choose proper datasets collection. This allows us to perform numerous experiments and rigorously analyze modern approaches. We arrive to many conclusions, which shed new light on performance and quality of novel algorithms. We investigate application of Jumping Knowledge GNN architecture to graph classification, which proves to be an efficient tool for improving base graph neural network architectures. Multiple improvements to baseline models are also proposed and experimentally verified, which constitutes an important contribution to the field of fair model comparison.
translated by 谷歌翻译
Most graph neural network models rely on a particular message passing paradigm, where the idea is to iteratively propagate node representations of a graph to each node in the direct neighborhood. While very prominent, this paradigm leads to information propagation bottlenecks, as information is repeatedly compressed at intermediary node representations, which causes loss of information, making it practically impossible to gather meaningful signals from distant nodes. To address this issue, we propose shortest path message passing neural networks, where the node representations of a graph are propagated to each node in the shortest path neighborhoods. In this setting, nodes can directly communicate between each other even if they are not neighbors, breaking the information bottleneck and hence leading to more adequately learned representations. Theoretically, our framework generalizes message passing neural networks, resulting in provably more expressive models, and we show that some recent state-of-the-art models are special instances of this framework. Empirically, we verify the capacity of a basic model of this framework on dedicated synthetic experiments, and on real-world graph classification and regression benchmarks, and obtain state-of-the-art results.
translated by 谷歌翻译
我们提出了一个分散的“Local2Global”的图形表示学习方法,即可以先用来缩放任何嵌入技术。我们的Local2Global方法首先将输入图分成重叠的子图(或“修补程序”)并独立地培训每个修补程序的本地表示。在第二步中,我们通过估计使用来自贴片重叠的信息的刚性动作的一组刚性运动来将本地表示将本地表示与全局一致的表示。 Local2Global相对于现有工作的关键区别特征是,在分布式训练期间无需经常昂贵的参数同步训练曲线的培训。这允许Local2Global缩放到大规模的工业应用,其中输入图甚至可能均不适合存储器,并且可以以分布式方式存储。我们在不同大小的数据集上应用Local2Global,并表明我们的方法在边缘重建和半监督分类上的规模和准确性之间实现了良好的权衡。我们还考虑异常检测的下游任务,并展示如何使用Local2Global在网络安全网络中突出显示异常。
translated by 谷歌翻译
图形神经网络(GNN)是具有无核数据的应用的有前途的方法。但是,具有数亿节点的大规模图上的培训GNN既是资源又是耗时的。与DNN不同,GNN通常具有更大的内存足迹,因此GPU内存能力和PCIE带宽是GNN培训中的主要资源瓶颈。为了解决此问题,我们提出分叉:一种图形量化方法,通过显着减少内存足迹和PCIE带宽要求来加速GNN训练,以便GNN可以充分利用GPU计算功能。我们的关键见解是,与DNN不同,GNN不太容易发生量化引起的输入特征的信息丢失。我们确定图形特征量化中的主要准确性影响因素,从理论上证明,分叉训练会收敛到网络,在该网络中,损失在未压缩网络的最佳损失的$ \ epsilon $之内。我们使用几种流行的GNN模型和数据集对分叉进行了广泛的评估,包括最大的公共图数据集MAG240M上的图形。结果表明,分叉达到30以上的压缩率,并在边际准确性损失的情况下提高了GNN训练速度200%-320%。特别是,分叉在一小时内仅使用四个GPU在MAG240M上的训练图来实现记录。
translated by 谷歌翻译
图形神经网络(GNN)在学习强大的节点表示中显示了令人信服的性能,这些表现在保留节点属性和图形结构信息的强大节点表示中。然而,许多GNNS在设计有更深的网络结构或手柄大小的图形时遇到有效性和效率的问题。已经提出了几种采样算法来改善和加速GNN的培训,但他们忽略了解GNN性能增益的来源。图表数据中的信息的测量可以帮助采样算法来保持高价值信息,同时消除冗余信息甚至噪声。在本文中,我们提出了一种用于GNN的公制引导(MEGUIDE)子图学习框架。 MEGUIDE采用两种新颖的度量:功能平滑和连接失效距离,以指导子图采样和迷你批次的培训。功能平滑度专为分析节点的特征而才能保留最有价值的信息,而连接失败距离可以测量结构信息以控制子图的大小。我们展示了MEGUIDE在多个数据集上培训各种GNN的有效性和效率。
translated by 谷歌翻译
Machine Unerning是在收到删除请求时从机器学习(ML)模型中删除某些培训数据的影响的过程。虽然直接而合法,但从划痕中重新训练ML模型会导致高计算开销。为了解决这个问题,在图像和文本数据的域中提出了许多近似算法,其中SISA是最新的解决方案。它将训练集随机分配到多个碎片中,并为每个碎片训练一个组成模型。但是,将SISA直接应用于图形数据可能会严重损害图形结构信息,从而导致的ML模型实用程序。在本文中,我们提出了Grapheraser,这是一种针对图形数据量身定制的新型机器学习框架。它的贡献包括两种新型的图形分区算法和一种基于学习的聚合方法。我们在五个现实世界图数据集上进行了广泛的实验,以说明Grapheraser的学习效率和模型实用程序。它可以实现2.06 $ \ times $(小数据集)至35.94 $ \ times $(大数据集)未学习时间的改进。另一方面,Grapheraser的实现最高62.5美元\%$更高的F1分数,我们提出的基于学习的聚合方法可达到高达$ 112 \%$ $ F1分数。 github.com/minchen00/graph-unlearning}。}。}
translated by 谷歌翻译
尽管在深度学习的其他应用领域中取得了非常深的架构,但流行的图神经网络是浅层模型。这降低了建模能力,并使模型无法捕获远程关系。浅设计的主要原因是过度平滑的,这导致节点状态随着深度的增加而变得更加相似。我们建立在GNNS和Pagerank之间的紧密联系的基础上,为此,个性化的Pagerank介绍了对个性化向量的考虑。通过这个想法,我们提出了个性化的Pagerank图神经网络(PPRGNN),该神经网络将图形卷积网络扩展到无限深度模型,该模型有机会将邻居聚集重置回每个迭代中的初始状态。我们引入了一个很好的解释调整,以重置重置并证明我们的方法与独特解决方案的收敛性,而无需放置任何限制,即使无限地进行了许多邻居聚集。与个性化的Pagerank一样,我们的结果不会过度光滑。在这样做的同时,在我们保持内存复杂性恒定的同时,时间复杂性保持线性,而与网络的深度无关,使其比较大图。我们从经验上展示了方法对各种节点和图形分类任务的有效性。在几乎所有情况下,PPRGNN优于可比较的方法。
translated by 谷歌翻译
图形神经网络(GNNS)由于图形数据的规模和模型参数的数量呈指数增长,因此限制了它们在实际应用中的效用,因此往往会遭受高计算成本。为此,最近的一些作品着重于用彩票假设(LTH)稀疏GNN,以降低推理成本,同时保持绩效水平。但是,基于LTH的方法具有两个主要缺点:1)它们需要对密集模型进行详尽且迭代的训练,从而产生了极大的训练计算成本,2)它们仅修剪图形结构和模型参数,但忽略了节点功能维度,存在大量冗余。为了克服上述局限性,我们提出了一个综合的图形渐进修剪框架,称为CGP。这是通过在一个训练过程中设计在训练图周期修剪范式上进行动态修剪GNN来实现的。与基于LTH的方法不同,提出的CGP方法不需要重新训练,这大大降低了计算成本。此外,我们设计了一个共同策略,以全面地修剪GNN的所有三个核心元素:图形结构,节点特征和模型参数。同时,旨在完善修剪操作,我们将重生过程引入我们的CGP框架,以重新建立修剪但重要的连接。提出的CGP通过在6个GNN体系结构中使用节点分类任务进行评估,包括浅层模型(GCN和GAT),浅但深度散发模型(SGC和APPNP)以及Deep Models(GCNII和RESGCN),总共有14个真实图形数据集,包括来自挑战性开放图基准的大规模图数据集。实验表明,我们提出的策略在匹配时大大提高了训练和推理效率,甚至超过了现有方法的准确性。
translated by 谷歌翻译