Addressing Heterophily in Node Classification with Graph Echo State Networks

Node classification tasks on graphs are addressed via fully-trained deep message-passing models that learn a hierarchy of node representations via multiple aggregations of a node's neighbourhood. While effective on graphs that exhibit a high ratio of intra-class edges, this approach poses challenges in the opposite case, i.e. heterophily, where nodes belonging to the same class are usually further apart. In graphs with a high degree of heterophily, the smoothed representations based on close neighbours computed by convolutional models are no longer effective. So far, architectural variations in message-passing models to reduce excessive smoothing or rewiring the input graph to improve longer-range message passing have been proposed. In this paper, we address the challenges of heterophilic graphs with Graph Echo State Network (GESN) for node classification. GESN is a reservoir computing model for graphs, where node embeddings are recursively computed by an untrained message-passing function. Our experiments show that reservoir models are able to achieve better or comparable accuracy with respect to most fully trained deep models that implement ad hoc variations in the architectural bias or perform rewiring as a preprocessing step on the input graph, with an improvement in terms of efficiency/accuracy trade-off. Furthermore, our analysis shows that GESN is able to effectively encode the structural relationships of a graph node, by showing a correlation between iterations of the recursive embedding function and the distribution of shortest paths in a graph.

PDF Abstract Neurocomputing 2023 PDF Neurocomputing 2023 Abstract

Results from the Paper


 Ranked #1 on Node Classification on genius (1:1 Accuracy metric)

     Get a GitHub badge
Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Node Classification Actor GESN Accuracy 34.56 ± 0.76 # 38
Node Classification arXiv-year GESN Accuracy 48.80 ± 0.22 # 11
Node Classification on Non-Homophilic (Heterophilic) Graphs Chameleon (48%/32%/20% fixed splits) GESN 1:1 Accuracy 77.05 ± 1.24 # 2
Node Classification Citeseer (48%/32%/20% fixed splits) GESN 1:1 Accuracy 74.51 ± 2.14 # 23
Node Classification Cora (48%/32%/20% fixed splits) GESN 1:1 Accuracy 86.04 ± 1.01 # 22
Node Classification on Non-Homophilic (Heterophilic) Graphs Cornell (48%/32%/20% fixed splits) GESN 1:1 Accuracy 81.14 ± 6.00 # 19
Node Classification genius GESN 1:1 Accuracy 91.72 ± 0.08 # 1
Node Classification on Non-Homophilic (Heterophilic) Graphs Penn94 GESN 1:1 Accuracy 80.29 ± 0.41 # 19
Node Classification PubMed (48%/32%/20% fixed splits) GESN 1:1 Accuracy 89.20 ± 0.34 # 14
Node Classification on Non-Homophilic (Heterophilic) Graphs Squirrel (48%/32%/20% fixed splits) GESN 1:1 Accuracy 73.56 ± 1.62 # 2
Node Classification on Non-Homophilic (Heterophilic) Graphs Texas (48%/32%/20% fixed splits) GESN 1:1 Accuracy 84.31 ± 4.44 # 13
Node Classification on Non-Homophilic (Heterophilic) Graphs twitch-gamers GESN 1:1 Accuracy 68.34 ± 0.86 # 1
Node Classification on Non-Homophilic (Heterophilic) Graphs Wisconsin (48%/32%/20% fixed splits) GESN 1:1 Accuracy 83.33 ± 3.81 # 18

Methods


No methods listed for this paper. Add relevant methods here