Diversity of augmented graphs.

<p>Similarities (↑ larger = more overlap) and complementary diversities (↑ larger = more diverse). Lower is better for .</p>

Saved in:

Bibliographic Details
Main Author:	Hossein Rafieizadeh (22676722) (author)
Other Authors:	Hadi Zare (20073000) (author), Mohsen Ghassemi Parsa (22676725) (author), Hocine Cherifi (8177628) (author)
Published:	2025
Subjects:	Cell Biology Science Policy Environmental Sciences not elsewhere classified Biological Sciences not elsewhere classified Information Systems not elsewhere classified intrusions across social reconstructions across views level contrastive learning dual contrastive learning across six benchmarks div >< p view discrepancies underutilized augmented graph views dcor improves auroc view discrepancies level contrast dual autoencoder augmented view specific information six datasets reduces auroc publicly available preserves fine physical domains performing non maximum gain leaving cross identifying threats financial fraud existing graph dcor reconstructs dcor ), contrasts reconstructions attributed networks attribute patterns
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items