Kernel-Based Semi-Supervised Learning Over Multilayer Graphs

Ioannidis, Vassilis N.; Traganitis, Panagiotis A.; Shen, Yu‐Chu; Giannakis, Georgios B.

doi:10.1109/spawc.2018.8445870

Cited by 11 publications

(13 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Consider a graph G = (V, E) with vertex set V and edge set E with respective cardinalities |V| = N and |E| = E. A set of nodal features is collected across time to form the set of endogenous vectors {x(t) ∈ R N } T t=1 . External (or exogenous) observables {ζ(t) ∈ R Q } T t=1 can be also available corresponding to e.g., extra [27], and snapshots or layers of one network [28].…”

Section: Modeling Interactions Over Graphsmentioning

confidence: 99%

Self-Driven Graph Volterra Models for Higher-Order Link Prediction

Coutiño

Karanikolas

Leus

et al. 2020

ICASSP 2020 - 2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

Self Cite

View full text Add to dashboard Cite

Link prediction is one of the core problems in network and data science with widespread applications. While predicting pairwise nodal interactions (links) in network data has been investigated extensively, predicting higher-order interactions (higher-order links) is still not fully understood. Several approaches have been advocated to predict such higher-order interactions, but no principled method has been put forth to tackle this challenge so far. Cross-fertilizing ideas from Volterra series and linear structural equation models, the present paper introduces self-driven graph Volterra models that can capture higher-order interactions among nodal observables available in networked data. The novel model is validated for the higher-order link prediction task using real interaction data from social networks.

show abstract

Section: Modeling Interactions Over Graphsmentioning

confidence: 99%

Self-Driven Graph Volterra Models for Higher-Order Link Prediction

Coutiño

Karanikolas

Leus

et al. 2020

ICASSP 2020 - 2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

Self Cite

View full text Add to dashboard Cite

show abstract

“…Consequently, generalizing the traditional single-layer to multilayer networks that organize the nodes into different groups, called layers,is well motivated. For kernel-based approaches for function reconstruction over multilayer graphs see also [30].…”

Section: Kernel Name Functionmentioning

confidence: 99%

“…To illustrate the benefits of employing different loss functions (30) and (32), we compare the performance of SP-GK and SP-GK( ) in the presence of outlying noise. Each sample f s is contaminated with Gaussian noise o s of large variance σ 2 o with probability p = 0.1.…”

Section: Multi-kernelmentioning

confidence: 99%

Kernel-Based Inference of Functions Over Graphs

Ioannidis

Νικολακόπουλος

et al. 2018

Adaptive Learning Methods for Nonlinear System Modeling

Self Cite

View full text Add to dashboard Cite

The study of networks has witnessed an explosive growth over the past decades with several ground-breaking methods introduced. A particularly interesting -and prevalent in several fields of study -problem is that of inferring a function defined over the nodes of a network. This work presents a versatile kernel-based framework for tackling this inference problem that naturally subsumes and generalizes the reconstruction approaches put forth recently by the signal processing on graphs community. Both the static and the dynamic settings are considered along with effective modeling approaches for addressing real-world problems. The herein analytical discussion is complemented by a set of numerical examples, which showcase the effectiveness of the presented techniques, as well as their merits related to state-of-the-art methods.

show abstract

“…In the particular case of social networks, each layer of the graph could capture a specific form of social interaction, such as friendship, family bonds, or coworker-ties [2]. Albeit their ubiquitous presence, development of SSL methods that account for multi-relational networks is only in its infancy, see, e.g., [1,3]. Related work.…”

Section: Introductionmentioning

confidence: 99%

A Recurrent Graph Neural Network for Multi-relational Data

Ioannidis

Marqués

Giannakis

2019

ICASSP 2019 - 2019 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

Self Cite

View full text Add to dashboard Cite

The era of "data deluge" has sparked the interest in graph-based learning methods in a number of disciplines such as sociology, biology, neuroscience, or engineering. In this paper, we introduce a graph recurrent neural network (GRNN) for scalable semisupervised learning from multi-relational data. Key aspects of the novel GRNN architecture are the use of multi-relational graphs, the dynamic adaptation to the different relations via learnable weights, and the consideration of graph-based regularizers to promote smoothness and alleviate over-parametrization. Our ultimate goal is to design a powerful learning architecture able to: discover complex and highly non-linear data associations, combine (and select) multiple types of relations, and scale gracefully with respect to the size of the graph. Numerical tests with real datasets corroborate the design goals and illustrate the performance gains relative to competing alternatives.

show abstract

Kernel-Based Semi-Supervised Learning Over Multilayer Graphs

Cited by 11 publications

References 15 publications

Self-Driven Graph Volterra Models for Higher-Order Link Prediction

Self-Driven Graph Volterra Models for Higher-Order Link Prediction

Kernel-Based Inference of Functions Over Graphs

A Recurrent Graph Neural Network for Multi-relational Data

Contact Info

Product

Resources

About