Learning Graphs From Data: A Signal Representation Perspective

Xiaowen Dong, Pascal Frossard
2019
Article

Résumé

The construction of a meaningful graph topology plays a crucial role in the effective representation, processing, analysis, and visualization of structured data. When a natural choice of the graph is not readily available from the data sets, it is thus desirable to infer or learn a graph topology from the data. In this article, we survey solutions to the problem of graph learning, including classical viewpoints from statistics and physics, and more recent approaches that adopt a graph signal processing (GSP) perspective. We further emphasize the conceptual similarities and differences between classical and GSP-based graph-inference methods and highlight the potential advantage of the latter in a number of theoretical and practical scenarios. We conclude with several open issues and challenges that are keys to the design of future signal processing and machine-learning algorithms for learning graphs from data.

Official source

https://infoscience.epfl.ch/record/265378?ln=fr

À propos de ce résultat

Cette page est générée automatiquement et peut contenir des informations qui ne sont pas correctes, complètes, à jour ou pertinentes par rapport à votre recherche. Il en va de même pour toutes les autres pages de ce site. Veillez à vérifier les informations auprès des sources officielles de l'EPFL.

Concepts associés

Chargement

Publications associées

Chargement

Robust and Efficient Data Clustering with Signal Processing on Graphs

Lionel Jérémie Martin

Data is pervasive in today's world and has actually been for quite some time. With the increasing volume of data to process, there is a need for faster and at least as accurate techniques than what we already have. In particular, the last decade recorded the effervescence of social networks and ubiquitous sensing (through smartphones and the Internet of Things). These phenomena, including also the progresses in bioinformatics and traffic monitoring, pushed forward the research on graph analysis and called for more efficient techniques. Clustering is an important field of machine learning because it belongs to the unsupervised techniques (i.e., one does not need to possess a ground truth about the data to start learning). With it, one can extract meaningful patterns from large data sources without requiring an expert to annotate a portion of the data, which can be very costly. However, the techniques of clustering designed so far all tend to be computationally demanding and have trouble scaling with the size of today's problems. The emergence of Graph Signal Processing, attempting to apply traditional signal processing techniques on graphs instead of time, provided additional tools for efficient graph analysis. By considering the clustering assignment as a signal lying on the nodes of the graph, one may now apply the tools of GSP to the improvement of graph clustering and more generally data clustering at large. In this thesis, we present several techniques using some of the latest developments of GSP in order to improve the scalability of clustering, while aiming for an accuracy resembling that of Spectral Clustering, a famous graph clustering technique that possess a solid mathematical intuition. On the one hand, we explore the benefits of random signal filtering on a practical and theoretical aspect for the determination of the eigenvectors of the graph Laplacian. In practice, this attempt requires the design of polynomial approximations of the step function for which we provided an accelerated heuristic. We used this series of work in order to reduce the complexity of dynamic graphs clustering, the problem of defining a partition to a graph which is evolving in time at each snapshot. We also used them to propose a fast method for the determination of the subspace generated by the first eigenvectors of any symmetrical matrix. This element is useful for clustering as it serves in Spectral Clustering but it goes beyond that since it also serves in graph visualization (with Laplacian Eigenmaps) and data mining (with Principal Components Projection). On the other hand, we were inspired by the latest works on graph filter localization in order to propose an extremely fast clustering technique. We tried to perform clustering by only using graph filtering and combining the results in order to obtain a partition of the nodes. These different contributions are completed by experiments using both synthetic datasets and real-world problems. Since we think that research should be shared in order to progress, all the experiments made in this thesis are publicly available on my personal Github account.

EPFL2018

Chargement

We live in a world characterized by massive information transfer and real-time communication. The demand for efficient yet low-complexity algorithms is widespread across different fields, including machine learning, signal processing and communications. Most of the problems that we encounter across these disciplines involves a large number of modules interacting with each other. It is therefore natural to represent these interactions and the flow of information between the modules in terms of a graph. This leads to the study of graph-based information processing framework. This framework can be used to gain insight into the development of algorithms for a diverse set of applications. We investigate the behaviour of large-scale networks (ranging from wireless sensor networks to social networks) as a function of underlying parameters. In particular, we study the scaling laws and applications of graph-based information processing in sensor networks/arrays, sparsity pattern recovery and interactive content search. In the first part of this thesis, we explore location estimation from incomplete information, a problem that arises often in wireless sensor networks and ultrasound tomography devices. In such applications, the data gathered by the sensors is only useful if we can pinpoint their positions with reasonable accuracy. This problem is particularly challenging when we need to infer the positions based on basic information/interaction such as proximity or incomplete (and often noisy) pairwise distances. As the sensors deployed in a sensor network are often of low quality and unreliable, we need to devise a mechanism to single out those that do not work properly. In the second part, we frame the network tomography problem as a well-studied inverse problem in statistics, called group testing. Group testing involves detecting a small set of defective items in a large population by grouping a subset of items into different pools. The result of each pool is a binary output depending on whether the pool contains a defective item or not. Motivated by the network tomography application, we consider the general framework of group testing with graph constraints. As opposed to conventional group testing where any subset of items can be grouped, here a test is admissible if it induces a connected subgraph. Given this constraint, we are interested in bounding the number of pools required to identify the defective items. Once the positions of sensors are known and the defective sensors are identified, we investigate another important feature of networks, namely, navigability or how fast nodes can deliver a message from one end to another by means of local operations. In the final part, we consider navigating through a database of objects utilizing comparisons. Contrary to traditional databases, users do not submit queries that are subsequently matched to objects. Instead, at each step, the database presents two objects to the user, who then selects among the pair the object closest to the target that she has in mind. This process continues until, based on the user’s answers, the database can identify the target she has in mind. The search through comparisons amounts to determining which pairs should be presented to the user in order to find the target object as quickly as possible. Interestingly, this problem has a natural connection with the navigability property studied in the second part, which enables us to develop efficient algorithms.

EPFL2012

Chargement

Laplacian Matrix Learning for Smooth Graph Signal Representation

Xiaowen Dong, Pascal Frossard, Pierre Vandergheynst

The construction of a meaningful graph plays a crucial role in the emerging field of signal processing on graphs. In this paper, we address the problem of learning graph Laplacians, which is similar to learning graph topologies, such that the input data form graph signals with smooth variations on the resulting topology. We adopt a factor analysis model for the graph signals and impose a Gaussian probabilistic prior on the latent variables that control these graph signals. We show that the Gaussian prior leads to an efficient representation that favours the smoothness property of the graph signals, and propose an algorithm for learning graphs that enforce such property. Experiments demonstrate that the proposed framework can efficiently infer meaningful graph topologies from only the signal observations.

2015

Chargement

EPFL2012

Robust and Efficient Data Clustering with Signal Processing on Graphs

Lionel Jérémie Martin

EPFL2018