Vertex-Context Sampling for Weighted Network Embedding

Network embedding methods have garnered increasing aŠention because of their e‚ectiveness in various information retrieval tasks. Œe goal is to learn low-dimensional representations of vertexes in an information network and simultaneously capture and preserve the network structure. Critical to the performance of a network embedding method is how the edges/vertexes of the network is sampled for the learning process. Many existing methods adopt a uniform sampling method to reduce learning complexity, but when the network is non-uniform (i.e. a weighted network) such uniform sampling incurs information loss. Œe goal of this paper is to present a generalized vertex sampling framework that works seamlessly with most existing network embedding methods to support weighted instead of uniform vertex/edge sampling. For eciency, we propose a delicate sequential vertex-to-context graph data structure, such that sampling a training pair for learning takes only constant time. For scalability and memory eciency, we design the graph data structure in a way that keeps space consumption low without requiring additional space. Moreover, the proposed framework can be used to implement extensions that feature high-order proximity modeling and weighted relation modeling. Experiments conducted on three datasets, including a commercial large-scale one, verify the e‚ectiveness and eciency of the proposedweighted network embedding methods on various tasks, including word similarity search, multi-label classi€cation, and item recommendation.