Multi-Hop Reasoning for Question Answering with Hyperbolic Representations
This work addresses the need for better modeling in knowledge graph-based reasoning, though it is incremental as it builds on existing hyperbolic representation methods.
The paper tackled the problem of multi-hop reasoning for question answering by comparing hyperbolic and Euclidean representations, finding that hyperbolic space consistently outperforms Euclidean space across diverse datasets, with learnable curvature initialization yielding superior results.
Hyperbolic representations are effective in modeling knowledge graph data which is prevalently used to facilitate multi-hop reasoning. However, a rigorous and detailed comparison of the two spaces for this task is lacking. In this paper, through a simple integration of hyperbolic representations with an encoder-decoder model, we perform a controlled and comprehensive set of experiments to compare the capacity of hyperbolic space versus Euclidean space in multi-hop reasoning. Our results show that the former consistently outperforms the latter across a diverse set of datasets. In addition, through an ablation study, we show that a learnable curvature initialized with the delta hyperbolicity of the utilized data yields superior results to random initializations. Furthermore, our findings suggest that hyperbolic representations can be significantly more advantageous when the datasets exhibit a more hierarchical structure.