论文阅读 [TPAMI-2022] PINE: Universal Deep Embedding for Graph Nodes via Partial Permutation Invariant Set Functions

论文搜索(studyai.com)

搜索论文: PINE: Universal Deep Embedding for Graph Nodes via Partial Permutation Invariant Set Functions

搜索论文: http://www.studyai.com/search/whole-site/?q=PINE:+Universal+Deep+Embedding+for+Graph+Nodes+via+Partial+Permutation+Invariant+Set+Functions

关键字(Keywords)

Task analysis; Laplace equations; Aggregates; Reinforcement learning; Matrix decomposition; Graph neural networks; Games; Graph embedding; partial permutation invariant set function; representation learning

机器学习; 自然语言处理; 强化学习; AI与Web

强化学习; 社区检测; 图网络; 图卷积网络; 语言表示学习; 矩阵因子分解

摘要(Abstract)

Graph node embedding aims at learning a vector representation for all nodes given a graph.

图节点嵌入旨在学习给定图的所有节点的向量表示。.

It is a central problem in many machine learning tasks (e.g., node classification, recommendation, community detection).

它是许多机器学习任务（例如节点分类、推荐、社区检测）中的核心问题。.

The key problem in graph node embedding lies in how to define the dependence to neighbors.

图节点嵌入的关键问题在于如何定义对邻域的依赖关系。.

Existing approaches specify (either explicitly or implicitly) certain dependencies on neighbors, which may lead to loss of subtle but important structural information within the graph and other dependencies among neighbors.

现有的方法（显式或隐式）指定了对邻域的某些依赖，这可能会导致丢失图中微妙但重要的结构信息以及邻域之间的其他依赖。.

This intrigues us to ask the question: can we design a model to give the adaptive flexibility of dependencies to each node’s neighborhood.

这激发了我们的兴趣，让我们提出这样一个问题：我们能否设计一个模型，为每个节点的邻域提供自适应的依赖灵活性。.

In this paper, we propose a novel graph node embedding method (named PINE) via a novel notion of partial permutation invariant set function, to capture any possible dependence.

在本文中，我们通过部分置换不变集函数的新概念，提出了一种新的图节点嵌入方法（称为PINE），以捕获任何可能的依赖关系。.

Our method 1) can learn an arbitrary form of the representation function from the neighborhood, without losing any potential dependence structures, and 2) is applicable to both homogeneous and heterogeneous graph embedding, the latter of which is challenged by the diversity of node types.

我们的方法1）可以从邻域中学习任意形式的表示函数，而不会丢失任何潜在的依赖结构；2）既适用于同质图嵌入，也适用于异构图嵌入，后者受到节点类型多样性的挑战。.

Furthermore, we provide theoretical guarantee for the representation capability of our method for general homogeneous and heterogeneous graphs.

此外，我们还为我们的方法对一般齐次图和非齐次图的表示能力提供了理论保证。.

Empirical evaluation results on benchmark data sets show that our proposed PINE method outperforms the state-of-the-art approaches on producing node vectors for various learning tasks of both homogeneous and heterogeneous graphs…

对基准数据集的实证评估结果表明，我们提出的PINE方法在为同质和异质图的各种学习任务生成节点向量方面优于最新的方法。。.