Conditional Structure Generation through Graph Variational Generative Adversarial Nets
Carl Yang · Peiye Zhuang · Wenhan Shi · Alan Luu · Pan Li

Thu Dec 12th 05:00 -- 07:00 PM @ East Exhibition Hall B + C #143

Graph embedding has been intensively studied recently, due to the advance of various neural network models. Theoretical analyses and empirical studies have pushed forward the translation of discrete graph structures into distributed representation vectors, but seldom considered the reverse direction, i.e., generation of graphs from given related context spaces. Particularly, since graphs often become more meaningful when associated with semantic contexts (e.g., social networks of certain communities, gene networks of certain diseases), the ability to infer graph structures according to given semantic conditions could be of great value. While existing graph generative models only consider graph structures without semantic contexts, we formulate the novel problem of conditional structure generation, and propose a novel unified model of graph variational generative adversarial nets (CondGen) to handle the intrinsic challenges of flexible context-structure conditioning and permutation-invariant generation. Extensive experiments on two deliberately created benchmark datasets of real-world context-enriched networks demonstrate the supreme effectiveness and generalizability of CondGen.

Author Information

Carl Yang (University of Illinois, Urbana Champaign)

Carl Yang is a final-year Ph.D. student with Jiawei Han in Computer Science at University of Illinois, Urbana Champaign. Before that, he received his B.Eng. in Computer Science at Zhejiang University under Xiaofei He in 2014. In his research, he develops data-driven techniques and neural architectures for learning with massive, complex and noisy graph data. His interests span data mining, machine learning and statistics, with a focus on leveraging graph analysis and deep learning techniques, to a wide range of questions including information network construction, entity/relation profiling, contextualized network embedding and so on. Carl’s leading-author research results have been published and well-cited in top conferences like KDD, WWW, NeurIPS, ICDE, ICDM, ECML-PKDD, CIKM, SDM and ICML.

Peiye Zhuang (UIUC)
Wenhan Shi (UIUC)
Alan Luu (UIUC)
Pan Li (Stanford)

More from the Same Authors