Online discriminative graph learning from multi-class smooth signals

  title={Online discriminative graph learning from multi-class smooth signals},
  author={Seyed Saman Saboksayr and Gonzalo Mateos and M{\"u}jdat Çetin},
  journal={Signal Process.},

Online Graph Learning under Smoothness Priors

This work develops novel algorithms for online network topology inference given streaming observations assumed to be smooth on the sought graph, and establishes that the online graph learning algorithm converges to within a neighborhood of the optimal time-varying batch solution.

EEG Emotion Recognition via Graph-based Spatio-Temporal Attention Neural Networks

The novel spatio-temporal attention neural network (STANN) is proposed to extract discriminative spatial and temporal features of EEG signals by a parallel structure of multi-column convolutional neural network and attention-based bidirectional long-short term memory.

Dual-based Online Learning of Dynamic Network Topologies

This work leverages and truncates dual-based proximal gradient iterations to solve a composite smoothness-regularized, time-varying inverse problem and shows that the online DPG algorithm converges faster than a primal-based baseline of comparable complexity.

Online Inference for Mixture Model of Streaming Graph Signals with Non-White Excitation

An expectation-maximization (EM) algorithm with a unique low-rank plus sparse prior derived from low pass signal property is designed and a novel online EM algorithm for inference from streaming data is proposed.

Simultaneous Graph Signal Clustering and Graph Learning

This paper addresses the problem of learning multiple graphs from heterogeneous data by formulating an optimization problem for joint graph signal clustering and graph topology inference and extends spectral clustering by partitioning the graph signals not only based on their pairwise similarities but also their smoothness with respect to the graphs associated with the clusters.

Graph Learning Based on Signal Smoothness Representation for Homogeneous and Heterogeneous Change Detection

A framework for CD based on graph fusion and driven by graph signal smoothness representation is proposed, which outperformed state-of-the-art approaches in ten out of 14 datasets.

Learning Time-Varying Graphs From Online Data

This work proposes an algorithmic framework to learn time-varying graphs from online data that is model-independent, i.e., it can be theoretically analyzed in its abstract formulation and then instantiated under a variety of model-dependent graph learning problems.

Online Graph Learning In Dynamic Environments

  • Xiang Zhang
  • Computer Science
    2022 30th European Signal Processing Conference (EUSIPCO)
  • 2022
Experimental results support that the proposed graph learning method is superior to the state-of-the-art methods, and dynamic regret analysis of the proposed method is performed, illustrating theoretically that proper dynamic priors do reduce regret.

Online Graph Learning From Time-Varying Structural Equation Models

This paper focuses on a time-varying version of the structural equation modeling (SEM) framework, which is an umbrella of multivariate techniques widely adopted in econometrics, epidemiology and psychology and views the linear SEM as a first-order diffusion of a signal over a graph whose topology changes over time.



Learning Laplacian Matrix in Smooth Graph Signal Representations

This paper addresses the problem of learning graph Laplacians, which is equivalent to learning graph topologies, such that the input data form graph signals with smooth variations on the resulting topology and proposes an algorithm for learning graphs that enforces such property and is based on minimizing the variations of the signals on the learned graph.

Disc-GLasso: Discriminative graph learning with sparsity regularization

This paper proposes an optimization framework for learning multiple graphs, each associated to a class of signals, such that representation of signals within a class and discrimination of signals in different classes are both taken into consideration.

Connecting the Dots: Identifying Network Structure via Graph Signal Processing

Challenges and opportunities for SP research in emerging topic areas at the crossroads of modeling, prediction, and control of complex behavior arising in networked systems that evolve over time are introduced.

Network Topology Inference from Spectral Templates

The novel idea is to find a graph shift that, while being consistent with the provided spectral information, endows the network with certain desired properties such as sparsity, and develops efficient inference algorithms stemming from provably tight convex relaxations of natural nonconvex criteria.

Online Graph Learning under Smoothness Priors

This work develops novel algorithms for online network topology inference given streaming observations assumed to be smooth on the sought graph, and establishes that the online graph learning algorithm converges to within a neighborhood of the optimal time-varying batch solution.

Graph Laplacian Mixture Model

This work proposes a novel generative model that represents a collection of distinct data which naturally live on different graphs that demonstrates promising performance in data clustering and multiple graph inference, and shows desirable properties in terms of interpretability and coping with high dimensionality on weather and traffic data.

Graph Topology Inference Based on Sparsifying Transform Learning

This paper considers two real-world applications of the proposed methods: the inference of the brain functional activity map from electrocorticography signals taken from patients affected by epilepsy, and the reconstruction of the radio environment map from sparse measurements of the electromagnetic field in an urban area.

Signal Processing on Directed Graphs: The Role of Edge Directionality When Processing and Learning From Network Data

A comprehensive review of recent advances in SP on digraphs is laid out, offering insights through comparisons with results available for undirected graphs, discussing emerging directions, and establishing links with related areas in machine learning and causal inference in statistics as well as illustrating their practical relevance to timely applications.

Topology Identification and Learning over Graphs: Accounting for Nonlinearities and Dynamics

The main goal of this paper is to outline overarching advances, and develop a principled framework to capture nonlinearities through kernels, which are judiciously chosen from a preselected dictionary to optimally fit the data.

Online Topology Inference from Streaming Stationary Graph Signals with Partial Connectivity Information

It is shown online proximal gradient iterations can be brought to bear to efficiently track the time-varying solution of the inverse problem with quantifiable guarantees and it is proved that the online algorithm converges to within a neighborhood of the optimal time-Varying batch solution.