TransH
Tags: #machine learning #KGEquation
$$f_{r}(h,t) =||h_{\perp} + d_{r} - t_{\perp} ||^{2}_{2}=||(h - w_{r}hw_{r}) + d_{r} - (t - w_{r}tw_{r}) ||^{2}_{2}$$Latex Code
f_{r}(h,t) =||h_{\perp} + d_{r} - t_{\perp} ||^{2}_{2}=||(h - w_{r}hw_{r}) + d_{r} - (t - w_{r}tw_{r}) ||^{2}_{2}
Have Fun
Let's Vote for the Most Difficult Equation!
Introduction
Equation
Latex Code
f_{r}(h,t) =||h_{\perp} + d_{r} - t_{\perp} ||^{2}_{2}=||(h - w_{r}hw_{r}) + d_{r} - (t - w_{r}tw_{r}) ||^{2}_{2}
Explanation
TransH model learns low-dimensional representations of knowledge graphs triples on the hyperplane of the entities and relations. See paper Knowledge Graph Embedding by Translating on Hyperplanes for more details.
Reply