BIDS Machine Learning and Science Forum
Date: Monday, May 3, 2021
Time: 11:00 AM - 12:00 PM Pacific Time
Location: Participate remotely using this Zoom link
HypE: Learning Knowledge Graph Entity Representations in Hyperbolic Space
Speaker: Nikhil Rao, Senior ML Scientist, Amazon
Abstract: Knowledge Graphs (KGs) are ubiquitous structures for information storage in several real-world applications such as web search, e-commerce, social networks, and biology. Querying KGs remains a foundational and challenging problem due to their size and complexity. State of the art methods embed representations of KG entities in Euclidean space, and perform some sort of matching to return answers to logical queries. In this talk, I will introduce the notion of embedding these entities in Hyperbolic space, which has been shown to better represent tree-like structures, which many Knowledge Graphs exhibit. The proposed representations can be trained in an end-to-end fashion by making modifications to neural models that learn Euclidean representations. On both KG question answering and e-commerce anomaly detection tasks, hyperbolic representations outperform several baselines.
The BIDS Machine Learning and Science Forum meets biweekly to discuss current applications across a wide variety of research domains in the physical sciences and beyond. These active sessions bring together domain scientists, statisticians, and computer scientists who are either developing state-of-the-art methods or are interested in applying these methods in their research. This Forum is organized by BIDS Faculty Affiliate Uroš Seljak (professor of Physics at UC Berkeley), BIDS Research Affiliate Ben Nachman (Physicist at Lawrence Berkeley National Laboratory), Vanessa Böhm and Ben Erichson. All interested members of the UC Berkeley and Berkeley Lab communities are welcome and encouraged to attend. To receive email notifications about upcoming meetings, or to request more information, please contact email@example.com.