Learning Long Range Spatio-Temporal Representations over Continuous Time Dynamic Graphs with State Space Models

Continuous-time dynamic graphs (CTDGs) provide a richer framework to capture fine-grained temporal patterns in evolving relational data. Long-range information propagation is a key challenge while learning representations, wherein it is important to retain and update information over long temporal horizons. Existing approaches restrict models to capture one-hop or local temporal neighborhoods and fail to capture multi-hop or global structural patterns. To mitigate this, we derive a parameter-efficient state-space modeling framework for continuous-time dynamic graphs (CTDG-SSM) from first principles. We first introduce continuous-time Topology-Aware higher order polynomial projection operator (CTT-HiPPO), a novel memory-based reformulation of HiPPO to jointly encode temporal dynamics and graph structure. The solution from CTT-HiPPO is obtained by projecting the classical HiPPO solution through a polynomial of the Laplacian matrix, yielding topology-aware memory updates that admit an equivalent state-space formulation for CTDGs (CTDG-SSM). Then a computationally efficient discrete formulation is obtained using the zero-order hold approach for model implementation. Across benchmarks on dynamic link prediction, dynamic node classification, and sequence classification, CTDG-SSM achieves state-of-the-art performance. Notably, it achieves large performance gains on datasets that require long range temporal (LRT) and spatial reasoning.

Task-driven Heterophilic Graph Structure Learning

Graph neural networks (GNNs) often struggle to learn discriminative node representations for heterophilic graphs, where connected nodes tend to have dissimilar labels and feature similarity provides weak structural cues. We propose frequency-guided graph structure learning (FgGSL), an end-to-end graph inference framework that jointly learns homophilic and heterophilic graph structures along with a spectral encoder. FgGSL employs a learnable, symmetric, feature-driven masking function to infer said complementary graphs, which are processed using pre-designed low- and high-pass graph filter banks. A label-based structural loss explicitly promotes the recovery of homophilic and heterophilic edges, enabling task-driven graph structure learning. We derive stability bounds for the structural loss and establish robustness guarantees for the filter banks under graph perturbations. Experiments on six heterophilic benchmarks demonstrate that FgGSL consistently outperforms state-of-the-art GNNs and graph rewiring methods, highlighting the benefits of combining frequency information with supervised topology inference.

Covariance Scattering Transforms

Machine learning and data processing techniques relying on covariance information are widespread as they identify meaningful patterns in unsupervised and unlabeled settings. As a prominent example, Principal Component Analysis (PCA) projects data points onto the eigenvectors of their covariance matrix, capturing the directions of maximum variance. This mapping, however, falls short in two directions: it fails to capture information in low-variance directions, relevant when, e.g., the data contains high-variance noise; and it provides unstable results in low-sample regimes, especially when covariance eigenvalues are close. CoVariance Neural Networks (VNNs), i.e., graph neural networks using the covariance matrix as a graph, show improved stability to estimation errors and learn more expressive functions in the covariance spectrum than PCA, but require training and operate in a labeled setup. To get the benefits of both worlds, we propose Covariance Scattering Transforms (CSTs), deep untrained networks that sequentially apply filters localized in the covariance spectrum to the input data and produce expressive hierarchical representations via nonlinearities. We define the filters as covariance wavelets that capture specific and detailed covariance spectral patterns. We improve CSTs' computational and memory efficiency via a pruning mechanism, and we prove that their error due to finite-sample covariance estimations is less sensitive to close covariance eigenvalues compared to PCA, improving their stability. Our experiments on age prediction from cortical thickness measurements on 4 datasets collecting patients with neurodegenerative diseases show that CSTs produce stable representations in low-data settings, as VNNs but without any training, and lead to comparable or better predictions w.r.t. more complex learning models.

Sampling and Recovery of Signals on a Simplicial Complex using Neighbourhood Aggregation

In this work, we focus on sampling and recovery of signals over simplicial complexes. In particular, we subsample a simplicial signal of a certain order and focus on recovering multi-order bandlimited simplicial signals of one order higher and one order lower. To do so, we assume that the simplicial signal admits the Helmholtz decomposition that relates simplicial signals of these different orders. Next, we propose an aggregation sampling scheme for simplicial signals based on the Hodge Laplacian matrix and a simple least squares estimator for recovery. We also provide theoretical conditions on the number of aggregations and size of the sampling set required for faithful reconstruction as a function of the bandwidth of simplicial signals to be recovered. Numerical experiments are provided to show the effectiveness of the proposed method.

Coded Aperture Radar Imaging Using Reconfigurable Intelligent Surfaces

In this paper, we focus on radar imaging using active sensing with a single transceiver and reconfigurable intelligent surface (RIS). RISs are arrays with tunable passive phase shifter elements that can modify the propagation channel. The RIS reflects each transmit pulse with a different phase profile. We use compressive sensing to recover the radar scene from observations at the single-antenna receiver. We also provide a projected gradient descent algorithm to design the RIS phase shifts to obtain minimally coherent observations required for recovery. Through numerical simulations, we demonstrate that the proposed method recovers radar scenes with point scatterers and extended targets.

Clustering with Simplicial Complexes

In this work, we propose a new clustering algorithm to group nodes in networks based on second-order simplices (aka filled triangles) to leverage higher-order network interactions. We define a simplicial conductance function, which on minimizing, yields an optimal partition with a higher density of filled triangles within the set while the density of filled triangles is smaller across the sets. To this end, we propose a simplicial adjacency operator that captures the relation between the nodes through second-order simplices. This allows us to extend the well-known Cheeger inequality to cluster a simplicial complex. Then, leveraging the Cheeger inequality, we propose the simplicial spectral clustering algorithm. We report results from numerical experiments on synthetic and real-world network data to demonstrate the efficacy of the proposed approach.

Learning to Precode for Integrated Sensing and Communications Systems

In this paper, we present an unsupervised learning neural model to design transmit precoders for integrated sensing and communication (ISAC) systems to maximize the worst-case target illumination power while ensuring a minimum signal-to-interference-plus-noise ratio (SINR) for all the users. The problem of learning transmit precoders from uplink pilots and echoes can be viewed as a parameterized function estimation problem and we propose to learn this function using a neural network model. To learn the neural network parameters, we develop a novel loss function based on the first-order optimality conditions to incorporate the SINR and power constraints. Through numerical simulations, we demonstrate that the proposed method outperforms traditional optimization-based methods in presence of channel estimation errors while incurring lesser computational complexity and generalizing well across different channel conditions that were not shown during training.

Optimal Placement of Active and Passive Elements in Hybrid RIS-assisted Communication Systems

Hybrid reconfigurable intelligent surfaces (HRIS) are RIS architectures having both active and passive elements. The received signal-to-noise ratio (SNR) in HRIS-assisted communication systems depends on the placement of active elements. In this paper, we show that received SNR can be improved with a channel-aware placement of the active elements. We jointly design the transmit precoder, the RIS coefficients, and the location of active and passive elements of the HRIS to maximize the SNR. We solve the underlying combinatorial nonconvex optimization problem using alternating optimization and propose a low-complexity solver, which is provably nearly optimal. Through numerical simulations, we demonstrate that the proposed method offers significantly improved performance compared to communication systems having a fully passive RIS array or a hybrid RIS array with channel agnostic active element placement and performance comparable to that of communication systems with a fully active RIS array.

Quantized Precoding and RIS-Assisted Modulation for Integrated Sensing and Communications Systems

In this paper, we present a novel reconfigurable intelligent surface (RIS)-assisted integrated sensing and communication (ISAC) system with 1-bit quantization at the ISAC base station. An RIS is introduced in the ISAC system to mitigate the effects of coarse quantization and to enable the co-existence between sensing and communication functionalities. Specifically, we design transmit precoder to obtain 1-bit sensing waveforms having a desired radiation pattern. The RIS phase shifts are then designed to modulate the 1-bit sensing waveform to transmit M-ary phase shift keying symbols to users. Through numerical simulations, we show that the proposed method offers significantly improved symbol error probabilities when compared to MIMO communication systems having quantized linear precoders, while still offering comparable sensing performance as that of unquantized sensing systems.

Integrated Sensing and Communications with Reconfigurable Intelligent Surfaces

Integrated sensing and communications (ISAC) are envisioned to be an integral part of future wireless networks, especially when operating at the millimeter-wave (mmWave) and terahertz (THz) frequency bands. However, establishing wireless connections at these high frequencies is quite challenging, mainly due to the penetrating pathloss that prevents reliable communication and sensing. Another emerging technology for next-generation wireless systems is reconfigurable intelligent surfaces (RISs), which are capable of modifying harsh propagation environments. RISs are the focus of growing research and industrial attention, bringing forth the vision of smart and programmable signal propagation environments. In this article, we provide a tutorial-style overview of the applications and benefits of RISs for sensing functionalities in general, and for ISAC systems in particular. We highlight the potential advantages when fusing these two emerging technologies, and identify for the first time that: i) joint sensing and communications designs are most beneficial when the channels referring to these operations are coupled, and that ii) RISs offer means for controlling this beneficial coupling. The usefulness of RIS-aided ISAC goes beyond the individual obvious gains of each of these technologies in both performance and power efficiency. We also discuss the main signal processing challenges and future research directions which arise from the fusion of these two emerging technologies.