HAN -

[trezbit]

Thank you for all this work!

In the book, Chapter 12, page 209, where a "Hierarchical self-Attention Network" (HAN) model was introduced to handle heterogeneous graphs, the reference [5] (J. Liu, Y. Wang, S. Xiang, and C. Pan. HAN: An Efficient Hierarchical Self-Attention Network for Skeleton-Based Gesture Recognition. arXiv, 2021. DOI: 10.48550/ARXIV.2106.13391. Available: https://arxiv.org/abs/2106.13391) points to a different architecture than the one actually implemented in Chapter 12.
Instead, all mathmetical presentations, down to the diagram illustrating 3-level architecture (node level, semantic level and prediction) seem to be directly borrowed from HAN for "Heterogeneous graph attention network": by Xiao Wang, Houye Ji, Chuan Shi, Bai Wang, Yanfang Ye, Peng Cui, and Philip S Yu. 2019.
Here is the actual paper: https://par.nsf.gov/servlets/purl/10135600,

Hope this helps.

[mlabonne]

Oh my bad, it's a mistake. You're correct, the citation is wrong and should reference the "Heterogeneous graph attention network" paper.