CEGRL-TKGR: A Causal Enhanced Graph Representation Learning Framework for Improving Temporal Knowledge Graph Extrapolation Reasoning
Preprint
Abstract
Temporal knowledge graph reasoning (TKGR) is increasingly gaining attention for its ability to extrapolate new events from historical data, thereby enriching the inherently incomplete temporal knowledge graphs. Existing graph-based representation learning frameworks have made significant strides in developing evolving representations for both entities and relational embeddings. Despite these achievements, there's a notable tendency in these models to inadvertently learn biased data representations and mine spurious correlations, consequently failing to discern the causal relationships between events. This often leads to incorrect predictions based on these false correlations. To address this, we propose an innovative causal enhanced graph representation learning framework for TKGR (named CEGRL-TKGR). This framework introduces causal structures in graph-based representation learning to unveil the essential causal relationships between events, ultimately enhancing task performance. Specifically, we first disentangle the evolutionary representations of entities and relations in a temporal graph sequence into two distinct components, namely causal representations and confounding representations. Then, drawing on causal intervention theory, we advocate the utilization of causal representations for predictions, aiming to mitigate the effects of erroneous correlations caused by confounding features, thus achieving more robust and accurate predictions. Finally, extensive experimental results on six benchmark datasets demonstrate the superior performance of our model in the link prediction task.Keywords
All Related Versions
This publication has 11 references indexed in Scilit:
- An Overview of Knowledge Graph Reasoning: Key Technologies and ApplicationsJournal of Sensor and Actuator Networks, 2022
- TLogic: Temporal Logical Rules for Explainable Link Forecasting on Temporal Knowledge GraphsProceedings of the AAAI Conference on Artificial Intelligence, 2022
- Dynamic Graph Neural Networks for Sequential RecommendationIEEE Transactions on Knowledge and Data Engineering, 2022
- Temporal Knowledge Graph Reasoning Based on Evolutional Representation Learning2021
- Learning from History: Modeling Temporal Knowledge Graphs with Sequential Copy-Generation NetworksProceedings of the AAAI Conference on Artificial Intelligence, 2021
- Dynamic Knowledge Graph based Multi-Event Forecasting2020
- Recurrent Event Network: Autoregressive Structure Inferenceover Temporal Knowledge Graphs2020
- Temporal Relational Ranking for Stock PredictionACM Transactions on Information Systems, 2019
- Learning Sequence Encoders for Temporal Knowledge Graph Completion2018
- Shallow Information Extraction for the knowledge Web2013