%0 Conference Paper %1 pan2023hge %A Pan, Jiaxin %A Nayyeri, Mojtaba %A Li, Yinan %A Staab, Steffen %B Thirty-eighth Conference on Artificial Intelligence, AAAI, 2024, Vancouver, Canada, February 22 – February 25, 2024, %D 2024 %K ac-opamin myown %T HGE: Embedding Temporal Knowledge Graphs in a Product Space of Heterogeneous Geometric Subspaces %X Temporal knowledge graphs represent temporal facts (s,p,o,τ) relating a subject s and an object o via a relation label p at time τ, where τ could be a time point or time interval. Temporal knowledge graphs may exhibit static temporal patterns at distinct points in time and dynamic temporal patterns between different timestamps. In order to learn a rich set of static and dynamic temporal patterns and apply them for inference, several embedding approaches have been suggested in the literature. However, as most of them resort to single underlying embedding spaces, their capability to model all kinds of temporal patterns was severely limited by having to adhere to the geometric property of their one embedding space. We lift this limitation by an embedding approach that maps temporal facts into a product space of several heterogeneous geometric subspaces with distinct geometric properties, i.e.\ Complex, Dual, and Split-complex spaces. In addition, we propose a temporal-geometric attention mechanism to integrate information from different geometric subspaces conveniently according to the captured relational and temporal information. Experimental results on standard temporal benchmark datasets favorably evaluate our approach against state-of-the-art models.

@inproceedings{pan2023hge, abstract = {Temporal knowledge graphs represent temporal facts (s,p,o,τ) relating a subject s and an object o via a relation label p at time τ, where τ could be a time point or time interval. Temporal knowledge graphs may exhibit static temporal patterns at distinct points in time and dynamic temporal patterns between different timestamps. In order to learn a rich set of static and dynamic temporal patterns and apply them for inference, several embedding approaches have been suggested in the literature. However, as most of them resort to single underlying embedding spaces, their capability to model all kinds of temporal patterns was severely limited by having to adhere to the geometric property of their one embedding space. We lift this limitation by an embedding approach that maps temporal facts into a product space of several heterogeneous geometric subspaces with distinct geometric properties, i.e.\ Complex, Dual, and Split-complex spaces. In addition, we propose a temporal-geometric attention mechanism to integrate information from different geometric subspaces conveniently according to the captured relational and temporal information. Experimental results on standard temporal benchmark datasets favorably evaluate our approach against state-of-the-art models.}, added-at = {2024-02-08T16:49:32.000+0100}, author = {Pan, Jiaxin and Nayyeri, Mojtaba and Li, Yinan and Staab, Steffen}, biburl = {https://puma.ub.uni-stuttgart.de/bibtex/2ad6aaef483f5761b15877266400bceeb/analyticcomp}, booktitle = {Thirty-eighth Conference on Artificial Intelligence, {AAAI}, 2024, Vancouver, Canada, February 22 – February 25, 2024,}, eventdate = {22 February 2024 – 25 February 2024}, interhash = {036f6272de90aabf31962bbf75072d25}, intrahash = {ad6aaef483f5761b15877266400bceeb}, keywords = {ac-opamin myown}, preprinturl = {https://arxiv.org/abs/2312.13680}, timestamp = {2024-02-08T16:53:47.000+0100}, title = {HGE: Embedding Temporal Knowledge Graphs in a Product Space of Heterogeneous Geometric Subspaces}, year = 2024 }