MöbiusE: Knowledge Graph Embedding on Möbius Ring

In this work, we propose a novel Knowledge Graph Embedding (KGE) strategy, called M\"{o}biusE, in which the entities and relations are embedded to surface of M\"{o}bius ring. The proposition such strategy is inspired by classic TorusE, addition two arbitrary elements subject modulus operation. sense, TorusE naturally guarantees critical boundedness embedding vectors KGE. However, nonlinear property operation on Torus ring uniquely derived operation, some extent restricts expressiveness TorusE. As further generalization M\"{o}biusE also uses preserve closeness it, but coordinates interacts with each other following way: {\em \color{red} any vector moves along its parametric trace will goes right opposite direction after cycle}. Hence, assumes much more representativeness than that turn it generates precise results. our experiments, outperforms strategies several key indicators.