no code implementations • 28 Mar 2024 • Jinghan Huang, Nanguang Chen, Anqi Qiu
This study, we introduce a novel Topological Cycle Graph Attention Network (CycGAT), designed to delineate a functional backbone within brain functional graph--key pathways essential for signal transmissio--from non-essential, redundant connections that form cycles around this core structure.
no code implementations • 11 Mar 2024 • Jinghan Huang, Qiufeng Chen, Yijun Bian, Pengli Zhu, Nanguang Chen, Moo K. Chung, Anqi Qiu
Additionally, we propose a pooling operator to coarsen $k$-simplices, combining features through simplicial attention mechanisms of self-attention and cross-attention via transformers and SP operators, capturing topological interconnections across multiple dimensions of simplices.
no code implementations • 18 Feb 2023 • Jinghan Huang, Moo K. Chung, Anqi Qiu
We introduce a generic formulation of spectral filters on heterogeneous graphs by introducing the $k-th$ Hodge-Laplacian (HL) operator.
no code implementations • 25 Jan 2023 • Yijun Bian, Kun Zhang, Anqi Qiu, Nanguang Chen
Furthermore, we investigate the properties of the proposed measure and propose first- and second-order oracle bounds to show that fairness can be boosted via ensemble combination with theoretical learning guarantees.
no code implementations • 26 Oct 2020 • Shih-Gu Huang, Moo K. Chung, Anqi Qiu, Alzheimer's Disease Neuroimaging Initiative
This paper revisits spectral graph convolutional neural networks (graph-CNNs) given in Defferrard (2016) and develops the Laplace-Beltrami CNN (LB-CNN) by replacing the graph Laplacian with the LB operator.
no code implementations • 6 Oct 2020 • Shih-Gu Huang, Moo K. Chung, Anqi Qiu, Alzheimer's Disease Neuroimaging Initiative
Even though graph convolutional neural network (graph-CNN) has been widely used in deep learning, there is a lack of augmentation methods to generate data on graphs or surfaces.
no code implementations • 7 Nov 2019 • Shih-Gu Huang, Ilwoo Lyu, Anqi Qiu, Moo. K. Chung
We also derive the closed-form expression of the spectral decomposition of the Laplace-Beltrami operator and use it to solve heat diffusion on a manifold for the first time.
no code implementations • 21 Mar 2019 • Caoqiang Liu, Hui Ji, Anqi Qiu
We developed a convolution neural network (CNN) on semi-regular triangulated meshes whose vertices have 6 neighbours.
no code implementations • 23 Sep 2014 • Moo. K. Chung, Anqi Qiu, Seongho Seo, Houri K. Vorperian
Starting with the heat kernel constructed from the eigenfunctions, we formulate a new bivariate kernel regression framework as a weighted eigenfunction expansion with the heat kernel as the weights.
no code implementations • 10 Oct 2013 • Jia Du, Alvina Goh, Anqi Qiu
We first assume that the atlas is generated from a known hyperatlas through a flow of diffeomorphisms and its shape prior can be constructed based on the framework of large deformation diffeomorphic metric mapping (LDDMM).
no code implementations • 25 Sep 2013 • Jia Du, A. Pasha Hosseinbor, Moo. K. Chung, Barbara B. Bendlin, Gaurav Suryawanshi, Andrew L. Alexander, Anqi Qiu
In this work, we show that the reorientation of the $q$-space signal due to spatial transformation can be easily defined on the BFOR signal basis.