Hypergraph neural networks (HGNNs) have shown great potential in modeling higher-order relationships among multiple entities. However, most existing HGNNs primarily emphasize low-pass filtering while neglecting the role of high-frequency information. In this work, we present a theoretical investigation into the spectral behavior of HGNNs and prove that combining both low-pass and high-pass components leads to more expressive and effective models. Notably, our analysis highlights that high-pass signals play a crucial role in capturing local discriminative structures within hypergraphs. Guided by these insights, we propose a novel sheaflet-based HNNs that integrates cellular sheaf theory and framelet transforms to preserve higher-order dependencies while enabling multi-scale spectral decomposition. This framework explicitly emphasizes high-pass components, aligning with our theoretical findings. Extensive experiments on benchmark datasets demonstrate the superiority of our approach over existing methods, validating the importance of high-frequency information in hypergraph learning.

Hypergraphs provide a natural and expressive framework for modeling high-order relationships, enabling the representation of group-wise interactions beyond pairwise connections. While hypergraph neural networks (HNNs) have shown promise for learning on such structures, existing models often rely on shallow message passing and lack the ability to extract multiscale patterns. Framelet-based techniques offer a principled solution by decomposing signals into multiple frequency bands. However, most prior framelet systems, particularly Haar-type ones, are sensitive to node ordering and fail to ensure consistent representations under permutation, leading to instability in hypergraph learning. To address this, we propose Permutation Equivariant Framelet-based Hypergraph Neural Networks (PEF-HNN), a novel framework that integrates multiscale framelet analysis with permutation-consistent learning. We construct a new family of permutation equivariant Haar-type framelets specifically designed for hypergraphs, supported by theoretical analysis of their stability and decomposition properties. Built upon these framelets, PEF-HNN incorporates both low-pass and high-pass components across multiple scales into a unified neural architecture. Extensive experiments on nine benchmark datasets, including three homophilic and four heterophilic hypergraphs, as well as two real-world datasets for visual object classification, demonstrate the effectiveness of our approach, consistently outperforming existing HNN baselines and highlighting the advantages of permutation equivariant framelet design in hypergraph representation learning.

Graph convolutional networks (GCNs) have emerged as powerful models for graph learning tasks, exhibiting promising performance in various domains. While their empirical success is evident, there is a growing need to understand their essential ability from a theoretical perspective. Existing theoretical research has primarily focused on the analysis of single-layer GCNs, while a comprehensive theoretical exploration of the stability and generalization of deep GCNs remains limited. In this paper, we bridge this gap by delving into the stability and generalization properties of deep GCNs, aiming to provide valuable insights by characterizing rigorously the associated upper bounds. Our theoretical results reveal that the stability and generalization of deep GCNs are influenced by certain key factors, such as the maximum absolute eigenvalue of the graph filter operators and the depth of the network. Our theoretical studies contribute to a deeper understanding of the stability and generalization properties of deep GCNs, potentially paving the way for developing more reliable and well-performing models.

Hypergraph neural networks (HNNs) have shown promise in handling tasks characterized by high-order correlations, achieving notable success across various applications. However, there has been limited focus on heterophilic hypergraph learning (HHL), in contrast to the increasing attention given to graph neural networks designed for graphs exhibiting heterophily. This paper aims to pave the way for HHL by addressing key gaps from multiple perspectives: measurement, dataset diversity, and baseline model development. First, we introduce metrics to quantify heterophily in hypergraphs, providing a numerical basis for assessing the homophily/heterophily ratio. Second, we develop diverse benchmark datasets across various real-world scenarios, facilitating comprehensive evaluations of existing HNNs and advancing research in HHL. Additionally, as a novel baseline model, we propose HyperUFG, a framelet-based HNN integrating both low-pass and high-pass filters. Extensive experiments conducted on synthetic and benchmark datasets highlight the challenges current HNNs face with heterophilic hypergraphs, while showcasing that HyperUFG performs competitively and often outperforms many existing models in such scenarios. Overall, our study underscores the urgent need for further exploration and development in this emerging field, with the potential to inspire and guide future research in HHL.

Focus measures are widely used to assess image clarity in various fields, such as photography and computer vision. However, many existing focus measures face challenges in balancing noise robustness and measurement capability. In this letter, a novel focus measure called Variance of Tight Framelet Feature (VTFF) is proposed to address this challenge. VTFF leverages the advantages of tight framelet features and variance information in feature maps to provide a robust and accurate assessment of image focus. Experimental results on both synthetic and real-world data demonstrate its superior performance compared to recent focus measures in measurement capability, noise robustness, and real-time performance.

Abstract. Each coil image in a parallel magnetic resonance imaging (pMRI) system is an imaging slice modulated by the corresponding coil sensitivity. These coil images, structurally similar to each other, are stacked together as 3-dimensional (3D) image data, and their sparsity property can be explored via 3D directional Haar tight framelets. The features of the 3D image data from the 3D framelet systems are utilized to regularize sensitivity encoding (SENSE) pMRI reconstruction. Accordingly, a so-called SENSE3d algorithm is proposed to reconstruct images of high quality from the sampled [Formula: see text]-space data with a high acceleration rate by decoupling effects of the desired image (slice) and sensitivity maps. Since both the imaging slice and sensitivity maps are unknown, this algorithm repeatedly performs a slice step followed by a sensitivity step by using updated estimations of the desired image and the sensitivity maps. In the slice step, for the given sensitivity maps, the estimation of the desired image is viewed as the solution to a convex optimization problem regularized by the sparsity of its 3D framelet coefficients of coil images. This optimization problem, involving data from the complex field, is solved by a primal-dual three-operator splitting (PD3O) method. In the sensitivity step, the estimation of sensitivity maps is modeled as the solution to a Tikhonov-type optimization problem that favors the smoothness of the sensitivity maps. This corresponding problem is nonconvex and could be solved by a forward-backward splitting method. Experiments on real phantoms and in vivo data show that the proposed SENSE3d algorithm can explore the sparsity property of the imaging slices and efficiently produce reconstructed images of high quality with reduced aliasing artifacts caused by high acceleration rate, additive noise, and the inaccurate estimation of each coil sensitivity. To provide a comprehensive picture of the overall performance of our SENSE3d model, we provide the quantitative index (HaarPSI) and comparisons to some deep learning methods such as VarNet and fastMRI-UNet.

In this article, we develop a general theoretical framework for constructing Haar-type tight framelets on any compact set with a hierarchical partition. In particular, we construct a novel area-regular hierarchical partition on the two spheres and establish its corresponding spherical Haar tight framelets with directionality. We conclude by evaluating and illustrate the effectiveness of our area-regular spherical Haar tight framelets in several denoising experiments. Furthermore, we propose a convolutional neural network (CNN) model for spherical signal denoising, which employs fast framelet decomposition and reconstruction algorithms. Experiment results show that our proposed CNN model outperforms threshold methods and processes strong generalization and robustness.

The nature of heterophilous graphs is significantly different from that of homophilous graphs, which causes difficulties in early graph neural network (GNN) models and suggests aggregations beyond the one-hop neighborhood. In this article, we develop a new way to implement multiscale extraction via constructing Haar-type graph framelets with desired properties of permutation equivariance, efficiency, and sparsity, for deep learning tasks on graphs. We further design a graph framelet neural network model permutation equivariant graph framelet augmented network (PEGFAN) based on our constructed graph framelets. The experiments are conducted on a synthetic dataset and nine benchmark datasets to compare the performance with other state-of-the-art models. The result shows that our model can achieve the best performance on certain datasets of heterophilous graphs (including the majority of heterophilous datasets with relatively larger sizes and denser connections) and competitive performance on the remaining.

Abstract. In this paper, we investigate in detail the structures of the variational characterization [Formula: see text] of the spherical [Formula: see text]-design, its gradient [Formula: see text], and its Hessian [Formula: see text] in terms of fast spherical harmonic transforms. Moreover, we propose solving the minimization problem of [Formula: see text] using the trust-region method to provide spherical [Formula: see text]-designs with large values of [Formula: see text]. Based on the obtained spherical [Formula: see text]-designs, we develop (semidiscrete) spherical tight framelets as well as their truncated systems and their fast spherical framelet transforms for the practical spherical signal/image processing. Thanks to the large spherical [Formula: see text]-designs and localization property of our spherical framelets, we are able to provide signal/image denoising using local thresholding techniques based on a fine-tuned spherical cap restriction. Many numerical experiments are conducted to demonstrate the efficiency and effectiveness of our spherical framelets and spherical designs, including Wendland function approximation, ETOPO data processing, and spherical image denoising.

Graph representation learning has many real-world applications, from self-driving LiDAR, 3D computer vision to drug repurposing, protein classification, social networks analysis. An adequate representation of graph data is vital to the learning performance of a statistical or machine learning model for graph-structured data. This paper proposes a novel multiscale representation system for graph data, called decimated framelets, which form a localized tight frame on the graph. The decimated framelet system allows storage of the graph data representation on a coarse-grained chain and processes the graph data at multi scales where at each scale, the data is stored on a subgraph. Based on this, we establish decimated G-framelet transforms for the decomposition and reconstruction of the graph data at multi resolutions via a constructive data-driven filter bank. The graph framelets are built on a chain-based orthonormal basis that supports fast graph Fourier transforms. From this, we give a fast algorithm for the decimated G-framelet transforms, or FGT, that has linear computational complexity O(N) for a graph of size N. The effectiveness for constructing the decimated framelet system and the FGT is demonstrated by a simulated example of random graphs and real-world applications, including multiresolution analysis for traffic network and representation learning of graph neural networks for graph classification tasks.

Deep Graph Neural Networks (GNNs) are useful models for graph classification and graph-based regression tasks. In these tasks, graph pooling is a critical ingredient by which GNNs adapt to input graphs of varying size and structure. We propose a new graph pooling operation based on compressive Haar transforms — \emph{HaarPooling}. HaarPooling implements a cascade of pooling operations; it is computed by following a sequence of clusterings of the input graph. A HaarPooling layer transforms a given input graph to an output graph with a smaller node number and the same feature dimension; the compressive Haar transform filters out fine detail information in the Haar wavelet domain. In this way, all the HaarPooling layers together synthesize the features of any given input graph into a feature vector of uniform size. Such transforms provide a sparse characterization of the data and preserve the structure information of the input graph. GNNs implemented with standard graph convolution layers and HaarPooling layers achieve state of the art performance on diverse graph classification and regression problems.

Compactly supported tight framelets are of great interest and importance in both theory and application. In this paper we discuss how to construct directional compactly supported tensor product complex tight framelets having varied directionality and good performance for applications in image processing. Our construction algorithms employ optimization techniques and put extensive emphasis on frequency response and spatial localization of their underlying one-dimensional tight framelet filter banks. Several concrete examples of directional compactly supported tensor product complex tight framelet filter banks are provided in this paper. Our numerical experiments show that such constructed directional compactly supported tensor product complex tight framelets have good performance for applications such as image denoising and inpainting compared with several other state-of-the-art transform-based methods.

Digital affine shear filter banks with 2-layer structure (DAS-2 filter banks) are constructed and are shown to be with the perfect reconstruction (PR) property. The implementation of digital affine shear transforms using the transition and subdivision operators are given. The redundancy rate analysis shows that our digital affine shear transforms have redundancy rate no more than 8 and it decreases with respect to the number of directional filters. Numerical experiments on image processing demonstrate the advantage of our DAS-2 filter banks over many other state-of-the-art frame-based transforms. The connection between DAS-2 filter banks and affine shear tight frames with 2-layer structure is established. Characterizations and constructions of affine shear tight frames with 2-layer structure are provided.

In this paper, we discuss the digitization and applications of smooth affine shear tight frames, a recently developed new class of directional multiscale representation systems. An affine wavelet tight frame is generated by isotropic dilations and translations of directional wavelet generators, while an affine shear tight frame is generated by anisotropic dilations, shears, and translations of shearlet generators. These two tight frames are actually connected in the sense that an affine shear tight frame can be obtained from an affine wavelet tight frame through subsampling. Consequently, an affine shear tight frame has an underlying filter bank from the MRA structure of its associated affine wavelet tight frame. We discuss the digitization of digital affine shear filter banks associated with the affine shear tight frames. Moreover, we provide the detailed algorithmic steps for both the forward and backward digital affine shear transforms. Analysis of the redundancy rate and computational complexity shows that the redundancy rate of the digital affine shear transforms does not increase with respect to the number of directions and the computational complexity is proportional to the redundancy rate and the FFT time for a fixed size of input data. Numerical experiments and comparisons in image/video processing show the advantages of our digital affine shear transforms over many other state-of-the-art frame-based directional transforms.

Multivariate problems are typically governed by anisotropic features such as edges in images. A common bracket of most of the various directional representation systems which have been proposed to deliver sparse approximations of such features is the utilization of parabolic scaling. One prominent example is the shearlet system. Our objective in this paper is threefold: We first develop a digital shearlet theory which is rationally designed in the sense that it is the digitization of the existing shearlet theory for continuous data. This implies that shearlet theory provides a unified treatment of both the continuum and digital realms. Second, we analyze the utilization of pseudo-polar grids and the pseudo-polar Fourier transform for digital implementations of parabolic scaling algorithms. We derive an isometric pseudo-polar Fourier transform by careful weighting of the pseudo-polar grid, allowing exploitation of its adjoint for the inverse transform. This leads to a digital implementation of the shearlet transform; an accompanying MATLAB toolbox called ShearLab (www.ShearLab.org) is provided. And, third, we introduce various quantitative measures for digital parabolic scaling algorithms in general, allowing one to tune parameters and objectively improve the implementation as well as compare different directional transform implementations. The usefulness of such measures is exemplarily demonstrated for the digital shearlet transform.