Despite the success of traditional network analysis, standard networks provide a limited representation of complex systems, which often include different types of relationships (or …
I Chami, Z Ying, C Ré… - Advances in neural …, 2019 - proceedings.neurips.cc
Graph convolutional neural networks (GCNs) embed nodes in a graph into Euclidean space, which has been shown to incur a large distortion when embedding real-world graphs with …
There has been a surge of recent interest in graph representation learning (GRL). GRL methods have generally fallen into three main categories, based on the availability of …
Q Liu, M Nickel, D Kiela - Advances in neural information …, 2019 - proceedings.neurips.cc
Learning from graph-structured data is an important task in machine learning and artificial intelligence, for which Graph Neural Networks (GNNs) have shown great promise. Motivated …
P Cui, X Wang, J Pei, W Zhu - IEEE transactions on knowledge …, 2018 - ieeexplore.ieee.org
Network embedding assigns nodes in a network to low-dimensional representations and effectively preserves the network structure. Recently, a significant amount of progresses …
M Nickel, D Kiela - Advances in neural information …, 2017 - proceedings.neurips.cc
Abstract Representation learning has become an invaluable approach for learning from symbolic data such as text and graphs. However, state-of-the-art embedding methods …
O Ganea, G Bécigneul… - Advances in neural …, 2018 - proceedings.neurips.cc
Hyperbolic spaces have recently gained momentum in the context of machine learning due to their high capacity and tree-likeliness properties. However, the representational power of …
AC Kozlowski, M Taddy… - American Sociological …, 2019 - journals.sagepub.com
We argue word embedding models are a useful tool for the study of culture using a historical analysis of shared understandings of social class as an empirical case. Word embeddings …
M Nickel, D Kiela - International conference on machine …, 2018 - proceedings.mlr.press
We are concerned with the discovery of hierarchical relationships from large-scale unstructured similarity scores. For this purpose, we study different models of hyperbolic …