On Modeling Hierarchical Data via Probabilistic Order Embeddings

By representing words with probability densities rather than point vectors, probabilistic word embeddings can capture rich and interpretable semantic information and uncertainty (Vilnis u0026 McCallum, 2014; Athiwaratkun u0026 Wilson, 2017). The uncertainty information can be particularly meaningful in capturing entailment relationships – whereby general words such as “entity” correspond to broad distributions that encompass more specific words such as “animal” or “instrument”. We introduce methodology to learn such representations effectively from labelled data. In particular, we propose simple yet effective loss functions and distance metrics, as well as graph-based schemes to select negative samples to better learn hierarchical probabilistic representations. Our approach provides state-of-the-art performance on the WordNet hypernym relationship prediction task and the challenging HyperLex lexical entailment dataset – while retaining a rich and interpretable probabilistic representation.