EvolveGCN: Evolving Graph Convolutional Networks for Dynamic Graphs

By Aldo Pareja, Giacomo Domeniconi, Tengfei Ma, Toyotaro Suzumura, H. Kanezashi, Tim Kaler, Charles E. Leisersen. 2019

1,524 citations recorded by the source corpus.

Graph representation learning resurges as a trending research subject owing to the widespread use of deep learning for Euclidean data, which inspire various creative designs of neural networks in the non-Euclidean domain, particularly graphs. With the success of these graph neural networks (GNN) in the static setting, we approach further practical scenarios where the graph dynamically evolves. Existing approaches typically resort to node embeddings and use a recurrent neural network (RNN, broadly speaking) to regulate the embeddings and learn the temporal dynamics. These methods require the knowledge of a node in the full time span (including both training and testing) and are less applicable to the frequent change of the node set. In some extreme scenarios, the node sets at different time steps may completely differ. To resolve this challenge, we propose EvolveGCN, which adapts the graph convolutional network (GCN) model along the temporal dimension without resorting to node embeddings. The proposed approach captures the dynamism of the graph sequence through using an RNN to evolve the GCN parameters. Two architectures are considered for the parameter evolution. We evaluate the proposed approach on tasks including link prediction, edge classification, and node classification. The experimental results indicate a generally higher performance of EvolveGCN compared with related approaches. The code is available at https://github.com/IBM/EvolveGCN.

Subfields: Artificial Intelligence, Cognitive Neuroscience, Statistical and Nonlinear Physics

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Cognitive Neuroscience

In Statistical and Nonlinear Physics

Citations in the corpus

Recurrent Event Network for Reasoning over Temporal Knowledge Graphs (2019, methodology): “EvolveGCN [18] and GCRN [21] introduce a RNN to update node representations across different time stamps.”
A Survey of Knowledge Graph Reasoning on Graph Types: Static, Dynamic, and Multi-Modal (2022, methodology): “EvolveGCN [182] adopts the graph convolutional networks (GCNs) to model the graph structure in each static snapshot and utilizes the LSTM model (also can utilize GRU model) to evolvetheGCNparametersovertime.”
Incremental Training of Graph Neural Networks on Temporal Graphs under Distribution Shift (2020, methodology): “EvolveGCN [19] learns a recurrent network to predict the parameters of a GCN for the next time step, such that vertex addition and removal is possible.”
Temporal Graph Networks for Deep Learning on Dynamic Graphs (2020, background): “Learning on dynamic graphs is relatively recent, and most works are limited to the setting of discrete-time dynamic graphs represented as a sequence of snapshots of the graph over time (37; 15; 68; 54; 48; 70).”
Foundations and Modeling of Dynamic Networks Using Dynamic Graph Neural Networks: A Survey (2020, methodology): “EvolveGCN [16] integrates an RNN into a GCN.”
Representation Learning for Dynamic Graphs: A Survey (2019, methodology): “Towards this goal, one approach is to concatenate the two vectors as [zv; zu] and feed the concatenation into a classifier (see, e.g., Pareja et al. (2019)).”
Spatio-Temporal Dynamic Inference Network for Group Activity Recognition (2021, background): “Some related works like EvolveGCN [30] explored a better representation learning strategy on evolving graphs and EvolveGraph [26] explored a prediction method to adjust the structure of one graph.”
Streaming Graph Neural Networks via Continual Learning (2020, methodology): “Some of them capture temporal information along with local structure so as to improve the expressive ability of themodel, with typical examples like DynamicTriad [34], HTNE [36], EPNE [27], EvolveGCN [18], DynamicGCN [4], DGNN [14] and ST-GCN [31].”
Inductive Representation Learning in Temporal Networks via Causal Anonymous Walks (2021, methodology): “We utilize EvolveGCN-O version since it can capture more graph structural information.”
Dynamic Hybrid Relation Network for Cross-Domain Context-Dependent Semantic Parsing (2021, methodology): “For each conversation that contains multiple turns, we design dynamic contextualized schema graphs, inspired by the recent studies on dynamically evolving structures (Pareja et al. 2020).”
Anti-Money Laundering in Bitcoin: Experimenting with Graph Convolutional Networks for Financial Forensics (2019, methodology): “The graph information may be instantiated in many ways; in EvolveGCN, it is represented by the embeddings of the top-k influential nodes in the graph.”
Measuring and Relieving the Over-smoothing Problem for Graph Neural Networks from the Topological View (2019, methodology): “Pareja et al. (2019) propose the EvolveGCN that uses the RNN to evolve the graph model itself over time.”
Recurrent Event Network: Autoregressive Structure Inference over Temporal Knowledge Graphs (2019, background): “seen in training, then their methods cannot predict events upon unseen time stamps. There is another line of works for Graph Convolutional Networks on temporal graphs such as (Seo et al., 2017), and (Pareja et al., 2019). Seo et al. (2017) proposed GCRN to model graph-structured sequence data. Their”
Relational Representation Learning for Dynamic (Knowledge) Graphs: A Survey (2019, methodology): “Towards this goal, one approach is to concatenate the two vectors as [ z v ; z u ] and feed the concatenation into a classiﬁer (see, e.g., [179]).”