Exploring the Limits of Historical Information for Temporal Knowledge Graph Extrapolation

TL;DR

This paper introduces CENET, a novel model leveraging contrastive learning to improve event forecasting in temporal knowledge graphs by effectively utilizing both historical and non-historical information.

Contribution

The paper proposes CENET, a new event forecasting model that incorporates contrastive learning to better handle non-recurrent events in temporal knowledge graphs.

Findings

CENET outperforms existing methods on five benchmark datasets.

Achieves at least 8.3% relative improvement of Hits@1 over baselines.

Effectively distinguishes between historical and non-historical dependencies.

Abstract

Temporal knowledge graphs, representing the dynamic relationships and interactions between entities over time, have been identified as a promising approach for event forecasting. However, a limitation of most temporal knowledge graph reasoning methods is their heavy reliance on the recurrence or periodicity of events, which brings challenges to inferring future events related to entities that lack historical interaction. In fact, the current state of affairs is often the result of a combination of historical information and underlying factors that are not directly observable. To this end, we investigate the limits of historical information for temporal knowledge graph extrapolation and propose a new event forecasting model called Contrastive Event Network (CENET) based on a novel training framework of historical contrastive learning. CENET learns both the historical and non-historical dependency to distinguish the most potential entities that best match the given query. Simultaneously, by launching contrastive learning, it trains representations of queries to probe whether the current moment is more dependent on historical or non-historical events. These representations further help train a binary classifier, whose output is a boolean mask, indicating the related entities in the search space. During the inference process, CENET employs a mask-based strategy to generate the final results. We evaluate our proposed model on five benchmark graphs. The results demonstrate that CENET significantly outperforms all existing methods in most metrics, achieving at least 8.3% relative improvement of Hits@1 over previous state-of-the-art baselines on event-based datasets.