Improving Few-Shot Multi-Hop Reasoning in Temporal Knowledge Graphs with Reinforcement Learning

Leszek Ziora

Improving Few-Shot Multi-Hop Reasoning in Temporal Knowledge Graphs with Reinforcement Learning

Authors: Leszek Ziora

Published: 2025-02-21 13:15:05.227

Abstract

This problem has unique challenges for few-shot multi-hop reasoning in TKGs, considering that the graph is dynamic and previous methods mainly focused on static graphs. In this paper, a reinforcement learning framework is integrated with advanced path search strategies to enhance the accuracy of reasoning, entity representation of tasks, and interpretability in TKGs. Five research hypotheses are considered: the effect of reinforcement learning, the contribution of one-hop neighbors, the efficacy of path search strategies, the relationship between the existing paths and the current state, and the contribution of path analysis to better interpretability. Quantitative methodologies are used with benchmark datasets, such as ICEWS18-few, ICEWS14-few, and GDELT-few. The results indicate that the framework improves the reasoning process and reduces computational complexity. These findings address the current gaps in TKG reasoning research and lay a foundation for advancing dynamic reasoning approaches in knowledge graphs.

Keywords

Few-shot reasoning, Temporal Knowledge Graphs, Reinforcement learning, Multi-hop reasoning

References

Bordes, A., Usunier, N., García-Durán, A., Weston, J., & Yakhnenko, O. (2013). Translating embeddings for modeling multi-relational data. Advances in Neural Information Processing Systems (NeurIPS), 26, 2787–2795
Das, R., Dhuliawala, S., Zaheer, M., Vilnis, L., Durme, B. V., & Smola, A. (2018). Go for a walk and arrive at the answer: Reasoning over paths in knowledge bases using reinforcement learning. Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing (EMNLP), 1932–1942
Ji, S., Pan, S., Cambria, E., Marttinen, P., & Yu, P. S. (2021). A survey on knowledge graphs: Representation, acquisition, and applications. IEEE Transactions on Neural Networks and Learning Systems, 33(2), 494–514
Lin, X. V., Socher, R., & Xiong, C. (2020). Multi-hop knowledge graph reasoning with reward shaping. Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP), 3243–3253
Trivedi, R., Dai, H., Wang, Y., & Song, L. (2017). Know-evolve: Deep temporal reasoning for dynamic knowledge graphs. Proceedings of the 34th International Conference on Machine Learning (ICML), 3462–3471
Xu, D., Radev, D. R., & Mohamed, A. M. (2020). Temporal knowledge graph completion using a linear time dynamic graph model. Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics (ACL), 4548–4558
Wang, Z., Mao, X., Wang, M., & Guo, L. (2014). Knowledge graph embedding by translating on hyperplanes. Proceedings of the Twenty-Eighth AAAI Conference on Artificial Intelligence (AAAI), 1112–1119
Yin, H., Sun, X., Li, Y., Chen, L., & Ma, J. (2020). Reinforcement learning-based temporal reasoning for dynamic knowledge graphs. IEEE Transactions on Knowledge and Data Engineering, 34(6), 2345–2358
Zhou, J., Wang, L., Chen, W., Wang, Y., & Yang, Z. (2022). Path-guided reasoning over temporal knowledge graphs with adaptive reinforcement learning. Proceedings of the 31st International Joint Conference on Artificial Intelligence (IJCAI), 3490–3496
Zhu, Z., & Sun, H. (2021). Few-shot reasoning over temporal knowledge graphs using meta-learning. Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics (ACL), 1691–1700

Download PDF

How to Cite

Leszek Ziora, (2025-02-21 13:15:05.227). Improving Few-Shot Multi-Hop Reasoning in Temporal Knowledge Graphs with Reinforcement Learning. Abhi International Journal of Information Processing Management, Volume oPMI31nYkkzgNQohcE9Z, Issue 1.

ISSN: XXXX-XXXX

Abhi International Journal of
Information Processing Management ( AIJIPM )