在处理机器学习模型（例如图形神经网络（GNN））中的一批图表时，通常将几个小图组合到一个整体图中以加速处理并减少填充的开销。例如，这是PYG库中支持的。但是，小图的尺寸对于节点和边缘的数量可能会有很大的变化，因此，组合图的大小仍然可能有很大差异，尤其是对于小批量大小而言。因此，仍然产生过多的填充和浪费计算的成本。本文提出了一种新方法 - 元组包装 - 用于生成导致最小开销的批次。该算法扩展了最近引入的序列填料方法，以在（| nodes |，| edges |）的2D元组上工作。单调启发式词被应用于元组值的2D直方图，以定义填充直方图箱的优先级，以及目标以达到节点数量和边缘数量的限制。实验验证了多个数据集上算法的有效性。

近年来，基于Weisfeiler-Leman算法的算法和神经架构，是一个众所周知的Graph同构问题的启发式问题，它成为具有图形和关系数据的机器学习的强大工具。在这里，我们全面概述了机器学习设置中的算法的使用，专注于监督的制度。我们讨论了理论背景，展示了如何将其用于监督的图形和节点表示学习，讨论最近的扩展，并概述算法的连接（置换 - ）方面的神经结构。此外，我们概述了当前的应用和未来方向，以刺激进一步的研究。

路由问题是许多实际应用的一类组合问题。最近，已经提出了端到端的深度学习方法，以了解这些问题的近似解决方案启发式。相比之下，经典动态编程（DP）算法保证最佳解决方案，但与问题大小严重规模。我们提出了深入的政策动态规划（DPDP），旨在将学习神经启发式的优势与DP算法结合起来。 DPDP优先确定并限制DP状态空间，使用来自深度神经网络的策略进行培训，以预测示例解决方案的边缘。我们在旅行推销员问题（TSP）上评估我们的框架，车辆路由问题（VRP）和TSP与时间窗口（TSPTW），并表明神经政策提高了（限制性）DP算法的性能，使其对强有力的替代品具有竞争力如LKH，同时也优于求解TSP，VRP和TSPTWS的大多数其他“神经方法”，其中包含100个节点。

我们提出了一个通用图形神经网络体系结构，可以作为任何约束满意度问题（CSP）作为末端2端搜索启发式训练。我们的体系结构可以通过政策梯度下降进行无监督的培训，以纯粹的数据驱动方式为任何CSP生成问题的特定启发式方法。该方法基于CSP的新型图表，既是通用又紧凑的，并且使我们能够使用一个GNN处理所有可能的CSP实例，而不管有限的Arity，关系或域大小。与以前的基于RL的方法不同，我们在全局搜索动作空间上运行，并允许我们的GNN在随机搜索的每个步骤中修改任何数量的变量。这使我们的方法能够正确利用GNN的固有并行性。我们进行了彻底的经验评估，从随机数据（包括图形着色，Maxcut，3-SAT和Max-K-Sat）中学习启发式和重要的CSP。我们的方法表现优于先验的神经组合优化的方法。它可以在测试实例上与常规搜索启发式竞争，甚至可以改善几个数量级，结构上比训练中看到的数量级更为复杂。

In recent years, methods based on deep neural networks, and especially Neural Improvement (NI) models, have led to a revolution in the field of combinatorial optimization. Given an instance of a graph-based problem and a candidate solution, they are able to propose a modification rule that improves its quality. However, existing NI approaches only consider node features and node-wise positional encodings to extract the instance and solution information, respectively. Thus, they are not suitable for problems where the essential information is encoded in the edges. In this paper, we present a NI model to solve graph-based problems where the information is stored either in the nodes, in the edges, or in both of them. We incorporate the NI model as a building block of hill-climbing-based algorithms to efficiently guide the election of neighborhood operations considering the solution at that iteration. Conducted experiments show that the model is able to recommend neighborhood operations that are in the $99^{th}$ percentile for the Preference Ranking Problem. Moreover, when incorporated to hill-climbing algorithms, such as Iterated or Multi-start Local Search, the NI model systematically outperforms the conventional versions. Finally, we demonstrate the flexibility of the model by extending the application to two well-known problems: the Traveling Salesman Problem and the Graph Partitioning Problem.

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.

近年来，在平衡（超级）图分配算法的设计和评估中取得了重大进展。我们调查了过去十年的实用算法的趋势，用于平衡（超级）图形分区以及未来的研究方向。我们的工作是对先前有关该主题的调查的更新。特别是，该调查还通过涵盖了超图形分区和流算法来扩展先前的调查，并额外关注并行算法。

\ textit {约束路径发现}的经典问题是一个经过充分研究但充满挑战的主题，在各个领域，例如沟通和运输等各个领域的应用。权重限制了最短路径问题（WCSPP），作为仅具有一个侧面约束的约束路径查找的基本形式，旨在计划成本最佳路径，其权重/资源使用受到限制。鉴于问题的双标准性质（即处理路径的成本和权重），解决WCSPP的方法具有一些带有双目标搜索的共同属性。本文在约束路径查找和双目标搜索中利用了最新的基于A*的最新技术，并为WCSPP提供了两种精确的解决方案方法，两者都可以在非常大的图表上解决硬性问题实例。我们从经验上评估了算法在新的大型和现实的问题实例上的性能，并在时空指标中显示出它们比最新算法的优势。本文还调查了优先级队列在被a*的约束搜索中的重要性。我们通过对逼真的和随机图进行了广泛的实验来展示，基于桶的队列没有打破打盘的方式可以有效地改善详尽的双标准搜索的算法性能。

快速扩大的神经网络模型在单个设备上运行越来越具有挑战性。因此，在多个设备上的模型并行性对于确保训练大型模型的效率至关重要。最近的建议在长时间处理时间或性能差。因此，我们提出了Celeritas，这是一个快速的框架，用于优化大型型号的设备放置。Celeritas在标准评估中采用简单但有效的模型并行化策略，并通过一系列调度算法生成位置策略。我们进行实验以在许多大型模型上部署和评估Celeritas。结果表明，与大多数高级方法相比，Celeritas不仅将放置策略生成时间减少26.4 \％，而且还将模型运行时间提高了34.2 \％。

本文研究了辍学图神经网络（DAVERGNNS），一种旨在克服标准GNN框架的局限性的新方法。在DAMPGNNS中，我们在输入图上执行多个GNN运行，其中一些节点随机且独立地在这些运行中丢弃。然后，我们将这些运行的结果结合起来获得最终结果。我们证明DAMPGNN可以区分无法通过GNN的消息分隔的各种图形邻域。我们导出了确保可靠分布辍学所需的运行数量的理论界限，我们证明了有关DACKGNNS的表现能力和限制的若干特性。我们在实验上验证了我们对表现力的理论结果。此外，我们表明DOWNNNS在已建立的GNN基准上表现得很竞争。

Supervised learning on molecules has incredible potential to be useful in chemistry, drug discovery, and materials science. Luckily, several promising and closely related neural network models invariant to molecular symmetries have already been described in the literature. These models learn a message passing algorithm and aggregation procedure to compute a function of their entire input graph. At this point, the next step is to find a particularly effective variant of this general approach and apply it to chemical prediction benchmarks until we either solve them or reach the limits of the approach. In this paper, we reformulate existing models into a single common framework we call Message Passing Neural Networks (MPNNs) and explore additional novel variations within this framework. Using MPNNs we demonstrate state of the art results on an important molecular property prediction benchmark; these results are strong enough that we believe future work should focus on datasets with larger molecules or more accurate ground truth labels.Recently, large scale quantum chemistry calculation and molecular dynamics simulations coupled with advances in high throughput experiments have begun to generate data at an unprecedented rate. Most classical techniques do not make effective use of the larger amounts of data that are now available. The time is ripe to apply more powerful and flexible machine learning methods to these problems, assuming we can find models with suitable inductive biases. The symmetries of atomic systems suggest neural networks that operate on graph structured data and are invariant to graph isomorphism might also be appropriate for molecules. Sufficiently successful models could someday help automate challenging chemical search problems in drug discovery or materials science.In this paper, our goal is to demonstrate effective machine learning models for chemical prediction problems

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}。}。}

在过去十年中，图形内核引起了很多关注，并在结构化数据上发展成为一种快速发展的学习分支。在过去的20年中，该领域发生的相当大的研究活动导致开发数十个图形内核，每个图形内核都对焦于图形的特定结构性质。图形内核已成功地成功地在广泛的域中，从社交网络到生物信息学。本调查的目标是提供图形内核的文献的统一视图。特别是，我们概述了各种图形内核。此外，我们对公共数据集的几个内核进行了实验评估，并提供了比较研究。最后，我们讨论图形内核的关键应用，并概述了一些仍有待解决的挑战。

Pre-publication draft of a book to be published byMorgan & Claypool publishers. Unedited version released with permission. All relevant copyrights held by the author and publisher extend to this pre-publication draft.

图形神经网络（GNNS）在学习从图形结构数据中展示了成功，其中包含欺诈检测，推荐和知识图形推理。然而，培训GNN有效地具有挑战性，因为：1）GPU存储器容量有限，对于大型数据集可能不足，而2）基于图形的数据结构导致不规则的数据访问模式。在这项工作中，我们提供了一种统计分析的方法，并确定了GNN培训前更频繁地访问的数据。我们的数据分层方法不仅利用输入图的结构，而且还从实际GNN训练过程中获得了洞察力，以实现更高的预测结果。通过我们的数据分层方法，我们还提供了一种新的数据放置和访问策略，以进一步最大限度地减少CPU-GPU通信开销。我们还考虑了多GPU GNN培训，我们也展示了我们在多GPU系统中的策略的有效性。评估结果表明，我们的工作将CPU-GPU流量降低了87-95％，并通过数亿节点和数十亿边缘的图表提高了现有解决方案的GNN训练速度。

In the last years, there has been a great interest in machine-learning-based heuristics for solving NP-hard combinatorial optimization problems. The developed methods have shown potential on many optimization problems. In this paper, we present a learned heuristic for the reoptimization of a problem after a minor change in its data. We focus on the case of the capacited vehicle routing problem with static clients (i.e., same client locations) and changed demands. Given the edges of an original solution, the goal is to predict and fix the ones that have a high chance of remaining in an optimal solution after a change of client demands. This partial prediction of the solution reduces the complexity of the problem and speeds up its resolution, while yielding a good quality solution. The proposed approach resulted in solutions with an optimality gap ranging from 0\% to 1.7\% on different benchmark instances within a reasonable computing time.

学习表达性分子表示对于促进分子特性的准确预测至关重要。尽管图形神经网络（GNNS）在分子表示学习中取得了显着进步，但它们通常面临诸如邻居探索，不足，过度光滑和过度阵列之类的局限性。同样，由于参数数量大，GNN通常具有较高的计算复杂性。通常，当面对相对大尺寸的图形或使用更深的GNN模型体系结构时，这种限制会出现或增加。克服这些问题的一个想法是将分子图简化为小型，丰富且有益的信息，这更有效，更具挑战性的培训GNN。为此，我们提出了一个新颖的分子图粗化框架，名为FUNQG利用函数组，作为分子的有影响力的构件来确定其性质，基于称为商图的图理论概念。通过实验，我们表明所产生的信息图比分子图小得多，因此是训练GNN的良好候选者。我们将FUNQG应用于流行的分子属性预测基准，然后比较所获得的数据集上的GNN体系结构的性能与原始数据集上的几个最先进的基线。通过实验，除了其参数数量和低计算复杂性的急剧减少之外，该方法除了其急剧减少之外，在各种数据集上的表现显着优于先前的基准。因此，FUNQG可以用作解决分子表示学习问题的简单，成本效益且可靠的方法。

Cardinality estimation is one of the most fundamental and challenging problems in query optimization. Neither classical nor learning-based methods yield satisfactory performance when estimating the cardinality of the join queries. They either rely on simplified assumptions leading to ineffective cardinality estimates or build large models to understand the data distributions, leading to long planning times and a lack of generalizability across queries. In this paper, we propose a new framework FactorJoin for estimating join queries. FactorJoin combines the idea behind the classical join-histogram method to efficiently handle joins with the learning-based methods to accurately capture attribute correlation. Specifically, FactorJoin scans every table in a DB and builds single-table conditional distributions during an offline preparation phase. When a join query comes, FactorJoin translates it into a factor graph model over the learned distributions to effectively and efficiently estimate its cardinality. Unlike existing learning-based methods, FactorJoin does not need to de-normalize joins upfront or require executed query workloads to train the model. Since it only relies on single-table statistics, FactorJoin has small space overhead and is extremely easy to train and maintain. In our evaluation, FactorJoin can produce more effective estimates than the previous state-of-the-art learning-based methods, with 40x less estimation latency, 100x smaller model size, and 100x faster training speed at comparable or better accuracy. In addition, FactorJoin can estimate 10,000 sub-plan queries within one second to optimize the query plan, which is very close to the traditional cardinality estimators in commercial DBMS.

决策树学习是机器学习中广泛使用的方法，在需要简洁明了的模型的应用中受到青睐。传统上，启发式方法用于快速生产具有相当高准确性的模型。然而，一个普遍的批评是，从精度和大小方面，所产生的树可能不一定是数据的最佳表示。近年来，这激发了最佳分类树算法的发展，这些算法与执行一系列本地最佳决策的启发式方法相比，在全球范围内优化决策树。我们遵循这一工作线，并提供了一种基于动态编程和搜索的最佳分类树的新颖算法。我们的算法支持对树的深度和节点数量的约束。我们方法的成功归因于一系列专门技术，这些技术利用了分类树独有的属性。传统上，最佳分类树的算法受到了高运行时的困扰和有限的可伸缩性，但我们在一项详细的实验研究中表明，我们的方法仅使用最先进的时间所需的时间，并且可以处理数十个数据集的数据集在数千个实例中，提供了几个数量级的改进，并特别有助于实现最佳决策树的实现。

The design of good heuristics or approximation algorithms for NP-hard combinatorial optimization problems often requires significant specialized knowledge and trial-and-error. Can we automate this challenging, tedious process, and learn the algorithms instead? In many real-world applications, it is typically the case that the same optimization problem is solved again and again on a regular basis, maintaining the same problem structure but differing in the data. This provides an opportunity for learning heuristic algorithms that exploit the structure of such recurring problems. In this paper, we propose a unique combination of reinforcement learning and graph embedding to address this challenge. The learned greedy policy behaves like a meta-algorithm that incrementally constructs a solution, and the action is determined by the output of a graph embedding network capturing the current state of the solution. We show that our framework can be applied to a diverse range of optimization problems over graphs, and learns effective algorithms for the Minimum Vertex Cover, Maximum Cut and Traveling Salesman problems.