About this document
Automated Hypergraph Neural Architecture Search by jwr15778002438 is a document available to read on EtoBox.
The paper introduces Hypergraph Neural Architecture Search (HyperNAS), a method for automatically designing optimal Hypergraph Neural Networks (HGNNs) to improve efficiency in handling non-Euclidean data. HyperNAS constructs a suitable search space for hypergraphs and employs differentiable search strategies, outperforming existing HGNNs and graph NAS methods in node classification tasks. Experimental results demonstrate its effectiveness on benchmark datasets, highlighting its potential for automatic hyper
- Author
- jwr15778002438
- Language
- EN