Abstract
Temporal Knowledge Graph (TKG) reasoning under the extrapolation setting focuses on forecasting future time-stamped events (facts) from historical data in a temporal knowledge graph. Existing approaches, particularly reinforcement learning (RL)-based multi-hop reasoning methods, are prominent for TKG reasoning as they produce human-interpretable predictions via explicit multi-hop path tracing. However, during RL training, rewards are typically sparse, and exploration is highly inefficient due to the vast, time-evolving action space. These issues hinder efficient training and often limit overall performance.
To address these challenges, we propose RAPTOR (Reachability-Aware Pretraining for Efficient Target-Oriented Path Exploration), a self-supervised pretraining method that injects a reachability-aware inductive bias to the agent. By learning to estimate the reachability of candidate actions to the target entity, RAPTOR reduces exploration over unpromising paths and provides a strong initialization for downstream RL fine-tuning.
Experimental results on the ICEWS14, ICEWS05-15, and ICEWS18 datasets demonstrate that RAPTOR pretraining markedly improves training efficiency and consistently outperforms conventional baselines, establishing it as an effective approach for enhancing RL-based multi-hop reasoning methods for TKG reasoning.
Blogger's Review: The introduction of RAPTOR presents a fresh perspective in TKG reasoning by incorporating reachability awareness, significantly enhancing training efficiency for multi-hop reasoning. This not only optimizes the exploration process but also lays a solid foundation for subsequent RL fine-tuning, making it worthwhile to validate its effectiveness in practical applications.