Abstract:Knowledge Graph (KG) reasoning has been an interesting topic in recent decades. Most current researches focus on predicting the missing facts for incomplete KG. Nevertheless, Temporal KG (TKG) reasoning, which is to forecast the future facts, still faces with the dilemma due to the complex interactions between entities over time. This paper proposes a novel intricate Spatiotemporal Dependency learning Network (STDN) based on Graph Convolutional Network (GCN) to capture the underlying correlations of an entity at different timestamps. Specifically, we first learn an adaptive adjacency matrix to depict the direct dependencies from the temporally adjacent facts of an entity, obtaining its previous context embedding. Then, a Spatiotemporal feature Encoding GCN (STE-GCN) is proposed to capture the latent spatiotemporal dependencies of the entity, getting the spatiotemporal embedding. Finally, a time gate unit is used to integrate the previous context embedding and the spatiotemporal embedding at the current timestamp to update the entity evolutional embedding for predicting future facts. STDN could generate the more expressive embeddings for capturing the intricate spatiotemporal dependencies in TKG. Extensive experiments on WIKI, ICEWS14 and ICEWS18 datasets prove our STDN has the advantage over state-of-the-art baselines for the temporal reasoning task.

What problem does this paper attempt to address?

The paper primarily focuses on addressing the problem of fact prediction in Temporal Knowledge Graphs (TKG), particularly the prediction of facts at future timestamps (i.e., extrapolation prediction). Specifically, the paper addresses the following issues: 1. **Complex Temporal Dependencies**: In temporal knowledge graphs, the relationships between entities change over time, often exhibiting complex patterns. Existing methods often struggle to capture these complex temporal dependencies. 2. **Simultaneously Occurring Facts**: Some methods can handle individual sequences of facts, but when multiple facts occur simultaneously (i.e., within the same time window), they may fail to effectively model these simultaneous facts. 3. **Long-term Dependencies**: Many methods use Recurrent Neural Networks (RNN) to capture the temporal evolution of entity embeddings, but they may lose important historical information in long sequences. To address the above challenges, the paper proposes a new Spatio-Temporal Dependency Learning Network (STDN) based on Graph Convolutional Network (GCN), aiming to capture deep spatio-temporal dependencies between entities at different timestamps. The specific steps are as follows: - **Adaptive Adjacency Matrix**: By learning an adaptive adjacency matrix to represent direct temporal dependencies, this helps capture the direct connections between entities at different time points. - **Spatio-Temporal Feature Encoding GCN (STE-GCN)**: This module is used to capture the latent spatio-temporal dependencies of entities, considering both spatial dependencies and dependencies at adjacent time points. - **Temporal Gating Unit (TGU)**: Used to integrate the previous contextual embeddings of entities with their spatio-temporal embeddings at the current timestamp to update the evolutionary embeddings of entities for future fact prediction. Experimental results show that the proposed STDN model outperforms existing static knowledge graph reasoning models and temporal knowledge graph reasoning models on three benchmark datasets (WIKI, ICEWS18, and ICEWS14) in the task of future fact prediction. Additionally, the model also demonstrates excellent performance in terms of prediction time and computational efficiency, being faster than other advanced whole-graph-based models. In summary, the main contribution of this paper is the proposal of a model that can effectively capture complex spatio-temporal dependencies, achieving significant improvements across multiple metrics.

Intricate Spatiotemporal Dependency Learning for Temporal Knowledge Graph Reasoning

CDRGN-SDE: Cross-Dimensional Recurrent Graph Network with neural Stochastic Differential Equation for temporal knowledge graph embedding

Learning Long- and Short-term Representations for Temporal Knowledge Graph Reasoning

Jointly Learning Time-Structure-History Graph Embedding for Temporal Knowledge Graph Reasoning.

Hierarchical Graph Attention Network for Temporal Knowledge Graph Reasoning.

Hierarchical Hypergraph Recurrent Attention Network for Temporal Knowledge Graph Reasoning

Temporal Knowledge Graph Reasoning Based on Dynamic Fusion Representation Learning

Temporal Inductive Path Neural Network for Temporal Knowledge Graph Reasoning

Spatiotemporal knowledge graph completion via diachronic and transregional word embedding

An Effective Time-Aware Encoder for Temporal Knowledge Graph Reasoning.

Temporal Knowledge Graph Reasoning Based on Graph Convolution Network

An embedding model for temporal knowledge graphs with long and irregular intervals

Spatial-Temporal Attention Network for Temporal Knowledge Graph Completion

Adaptive Path-Memory Network for Temporal Knowledge Graph Reasoning

Temporal knowledge graph reasoning triggered by memories

Learning Latent Relations for Temporal Knowledge Graph Reasoning

Complex Evolutional Pattern Learning for Temporal Knowledge Graph Reasoning

EvoKG: Jointly Modeling Event Time and Network Structure for Reasoning over Temporal Knowledge Graphs

Spatial-Temporal Knowledge Graph Network for Event Prediction

Temporal Knowledge Graph Reasoning Based on Evolutional Representation and Contrastive Learning