Track papers across labs, topics, and trends.
Search papers, labs, and topics across Lattice.
This paper introduces RAPTOR, a self-supervised pretraining method designed to enhance reinforcement learning (RL) for temporal knowledge graph (TKG) reasoning by incorporating a reachability-aware inductive bias. By estimating the reachability of candidate actions to target entities, RAPTOR significantly reduces inefficient exploration in the vast action space, leading to improved training efficiency. Experimental evaluations on multiple datasets show that RAPTOR consistently outperforms traditional baselines, marking a substantial advancement in the effectiveness of RL-based multi-hop reasoning methods for TKG tasks.
RAPTOR transforms TKG reasoning by drastically improving exploration efficiency, leading to faster and more accurate predictions in complex temporal environments.
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, reinforcement learning (RL)-based multi-hop reasoning methods are prominent for TKG reasoning because 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 the training efficiency and consistently outperforms conventional baselines, establishing it as an effective approach for enhancing RL-based multi-hop reasoning methods for TKG reasoning.