English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Graph neural networks (GNNs) are popular weapons for modeling relationaldata. Existing GNNs are not specified for attribute-incomplete graphs, makingmissing attribute imputation a burning issue. Until recently, many works noticethat GNNs are coupled with spectral concentration, which means the spectrumobtained by GNNs concentrates on a local part in spectral domain, e.g.,low-frequency due to oversmoothing issue. As a consequence, GNNs may beseriously flawed for reconstructing graph attributes as graph spectralconcentration tends to cause a low imputation precision. In this work, wepresent a regularized graph autoencoder for graph attribute imputation, namedMEGAE, which aims at mitigating spectral concentration problem by maximizingthe graph spectral entropy. Notably, we first present the method for estimatinggraph spectral entropy without the eigen-decomposition of Laplacian matrix andprovide the theoretical upper error bound. A maximum entropy regularizationthen acts in the latent space, which directly increases the graph spectralentropy. Extensive experiments show that MEGAE outperforms all the otherstate-of-the-art imputation methods on a variety of benchmark datasets.

Handling Missing Data via Max-Entropy Regularized Graph Autoencoder