[paper reading]Innovations in Graph Representation Learning

Google AI 原文链接

Given the widespread prevalence of graphs, graph analysis plays a fundamental role in machine learning, with applications in clustering, link prediction, privacy, and others.（图分析应用广泛）

However, graphs are inherently combinatorial structures made of discrete parts like nodes and edges, while many common ML methods, like neural networks, favor continuous structures, in particular vector representations. （图有其内在的离散结构信息，而CNN一般是连续信息。）

To get around the difficulties in using discrete graph representations in ML, graph embedding methods learn a continuous vector space for the graph, assigning each node (and/or edge) in the graph to a specific position in a vector space. (使用graph embedding，学习embedding空间，使每个node在其中可以用vector标识，从而由图结构转化为vector的形式)

注意其中node的向量化分布显示出了node之间的远近关系

In virtually all cases, the crucial assumption of standard graph embedding methods is that a single embedding has to be learned for each node. Thus, the embedding method can be said to seek to identify the single role or position that characterizes each node in the geometry of the graph. （之前的grapg embedding方法都是假设一个node只能学习到一个embedding？google提出了如下方法，一个node可以有多个embedding表示）

For instance, in the figure below, node U belongs to two communities: Cluster 1 (with the friends A and B, say U’s family members) and Cluster 2 (with C and D, say U’s colleagues).

Then, we use this information to “split” node U into its two personas U1 (the family persona) and U2 (the work persona). This disentangles the two communities, so that they no longer overlap.

Top Left: A typical graphs with highly overlapping communities. Top Right: A traditional embedding of the graph on the left using node2vec. Bottom Left: A persona graph of the graph above. Bottom Right: The Splitter embedding of the persona graph. Notice how the persona graph clearly disentangles the overlapping communities of the original graph and Splitter outputs well-separated embeddings.

由此可见，persona graph的效果