Table 1:

Notations and definitions

Notation	Definition
$𝒢=(𝒱,ℰ,\mathbf{w})$	Hypergraph
$𝒱\in {\mathbb{R}}^N$	The vertex set of the hypergraph
$ℰ\in {\mathbb{R}}^M$	The hyperedge set of the hypergraph
$\mathbf{H}\in {\mathbb{R}}^{N\times M}$	The incidence matrix of the hypergraph
${\mathbf{D}}_v\in {\mathbb{R}}^{N\times N}$	The diagonal vertex degree matrix
${\mathbf{D}}_e\in {\mathbb{R}}^{M\times M}$	The diagonal hyperedge degree matrix
$\mathbf{W}\in {\mathbb{R}}^{M\times M}$	The diagonal hyperedge weight matrix
$\mathbf{S}\in {\mathbb{R}}^{N\times N}$	The hypergraph similarity matrix
$\mathbf{L}\in {\mathbb{R}}^{N\times N}$	The hypergraph Laplacian matrix
$\mathbf{F}\in {\mathbb{R}}^{N\times P}$	The fMRI time series data
$\mathbf{U}\in {\mathbb{R}}^{N\times N}$	The eigenvector matrix of $\mathbf{L}$
${\mathbf{X}}^{\left(l\right)}\in {\mathbb{R}}^{N\times k^l}$	The hypergraph signal at l-th layer $({\mathbf{X}}^{\left(0\right)}=\mathbf{S})$
${\mathrm{\Theta }}^{\left(l\right)}\in {\mathbb{R}}^{k^{\left(l\right)}\times k^{(l+1)}}$	The learnable weight matrix at l-th layer
${\mathbf{A}}^l\in {\mathbb{R}}^{N\times K}$	The feature embedding at the last layer