Unsupervised Inductive Graph-Level Representation Learning via Graph-Graph Proximity

We introduce a novel approach to graph-level representation learning, which is to embed an entire graph into a vector space where the embeddings of two graphs preserve their graph-graph proximity. Our approach, UGRAPHEMB, is a general framework that provides a novel means to performing graph-level embedding in a completely unsupervised and inductive manner... (read more)

PDF Abstract

Datasets


Results from the Paper


TASK DATASET MODEL METRIC NAME METRIC VALUE GLOBAL RANK RESULT BENCHMARK
Graph Classification IMDb-M UGraphEmb-F Accuracy 50.97% # 12
Graph Classification IMDb-M UGraphEmb Accuracy 50.06% # 16
Graph Classification NCI109 UGraphEmb Accuracy 69.17 # 17
Graph Classification NCI109 UGraphEmb-F Accuracy 74.48 # 12
Graph Classification PTC UGraphEmb Accuracy 72.54% # 7
Graph Classification PTC UGraphEmb-F Accuracy 73.56% # 4
Graph Classification REDDIT-MULTI-12K UGraphEmb-F Accuracy 41.84 # 2
Graph Classification REDDIT-MULTI-12K UGraphEmb Accuracy 39.97 # 3
Graph Classification Web UGraphEmb-F Accuracy 45.03 # 1

Methods used in the Paper


METHOD TYPE
🤖 No Methods Found Help the community by adding them if they're not listed; e.g. Deep Residual Learning for Image Recognition uses ResNet