Corpus ID: 4475548

Signed Laplacian for spectral clustering revisited

@article{Knyazev2017SignedLF,
  title={Signed Laplacian for spectral clustering revisited},
  author={Andrew V. Knyazev},
  journal={ArXiv},
  year={2017},
  volume={abs/1701.01394}
}
  • A. Knyazev
  • Published 5 January 2017
  • Mathematics, Computer Science
  • ArXiv
Classical spectral clustering is based on a spectral decomposition of a graph Laplacian, obtained from a graph adjacency matrix representing positive graph edge weights describing similarities of graph vertices. In signed graphs, the graph edge weights can be negative to describe disparities of graph vertices, for example, negative correlations in the data. Negative weights lead to possible negative spectrum of the standard graph Laplacian, which is cured by defining a signed Laplacian. We… Expand
Spectral Clustering with Sampling — A Graph Signal Processing Perspective
Graph signal processing is an extension of classical signal processing to signals that are defined over the vertex set of graphs. This type of signal processing is necessary in applications where theExpand
Distance Preserving Model Order Reduction of Graph-Laplacians and Cluster Analysis
TLDR
Two novel algorithms are developed that ensure that target subset clustering is consistent with the spectral clustering of the full data set if one would perform such and a properly parametrized reduced-order model of the graph-Laplacian that approximates accurately the diffusion transfer function of the original graph for inputs and outputs restricted to the target subset. Expand
Node classification for signed networks using diffuse interface methods
TLDR
A recently introduced technique based on a partial differential equation defined on the signed network, namely the Allen-Cahn-equation, is extended to classify the nodes of the graph into two or more classes. Expand
Clustering of graph vertex subset via Krylov subspace model reduction
TLDR
Two novel algorithms for spectral clustering of a subset of the graph vertices (target subset) based on the theory of model order reduction rely on realizations of a reduced order model (ROM) that accurately approximates the diffusion transfer function of the original graph for inputs and outputs restricted to the target subset. Expand
Robust Semisupervised Graph Classifier Learning With Negative Edge Weights
TLDR
This paper constructs a complete graph-based binary classifier given only samples’ feature vectors and partial labels and proposes an algorithm based on iterative reweighted least squares that solves the posed MAP problem efficiently. Expand
Preconditioned spectral clustering for stochastic block partition streaming graph challenge (Preliminary version at arXiv.)
TLDR
Locally Optimal Block Preconditioned Conjugate Gradient (LOBPCG) is demonstrated to efficiently solve eigenvalue problems for graph Laplacians that appear in spectral clustering to achieve 100% correctness for all Challenge datasets with known truth partitions. Expand
Inhomogeneous Hypergraph Clustering with Applications
TLDR
It is proved that inhomogenous partitioning produces a quadratic approximation to the optimal solution if the inhomogeneous costs satisfy submodularity constraints and offers significant performance improvements in applications such as structure learning of rankings, subspace segmentation and motif clustering. Expand
Hypergraph clustering with inhomogeneous partitions of hyperedges Pan
The extension of conventional clustering to hypergraph clustering, which involves higher order relations among nodes instead of pairwise relations, is increasingly gaining attention, as multi-nodeExpand
Disc-GLasso: Discriminative graph learning with sparsity regularization
TLDR
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. Expand
Proximity Preserving Binary Code using Signed Graph-Cut
We introduce a binary embedding framework, called Proximity Preserving Code (PPC), which learns similarity and dissimilarity between data points to create a compact and affinity-preserving binaryExpand
...
1
2
...

References

SHOWING 1-10 OF 49 REFERENCES
Spectral Analysis of Signed Graphs for Clustering, Prediction and Visualization
TLDR
It is shown that several characteristic matrices of graphs can be extended to graphs with positively and negatively weighted edges, giving signed spectral clustering methods, signed graph kernels and network visualization methods that apply to signed graphs. Expand
Spectral Theory of Unsigned and Signed Graphs. Applications to Graph Clustering: a Survey
TLDR
A fairly thorough treatment of the method of normalized graph cuts, a deeply original method due to Shi and Malik, including complete proofs, is provided. Expand
On Spectral Analysis of Signed and Dispute Graphs
TLDR
This paper develops a clustering method based on matrix perturbation theorem for analyzing community structures of complex signed networks and proves and demonstrates cluster orthogonality for two types of signed networks: graph with dense inter-community mixed sign edges and k-dispute graph. Expand
PARTITIONING SPARSE MATRICES WITH EIGENVECTORS OF GRAPHS*
TLDR
It is shown that lower bounds on separator sizes can be obtained in terms of the eigenvalues of the Laplacian matrix associated with a graph, which can be used to compute good separators in grid graphs. Expand
A tutorial on spectral clustering
TLDR
This tutorial describes different graph Laplacians and their basic properties, present the most common spectral clustering algorithms, and derive those algorithms from scratch by several different approaches. Expand
First-principles multiway spectral partitioning of graphs
TLDR
This paper starts with a matrix formulation of the minimum cut problem and then shows, via a relaxed optimization, how it can be mapped onto a spectral embedding defined by the leading eigenvectors of the graph Laplacian, an algorithm that outperforms previous spectral partitioning approaches. Expand
Simple and Scalable Constrained Clustering: a Generalized Spectral Method
TLDR
A simple spectral approach to the well-studied constrained clustering problem with graph Laplacians that works in nearly-linear time and provides concrete guarantees for the quality of the clusters, at least for the case of 2-way partitioning. Expand
Spectral Theory for Dynamics on Graphs Containing Attractive and Repulsive Interactions
TLDR
This paper shows that the there are topological constraints on the index of the Laplacian matrix related to the dimension of a certain homology group, which gives bounds on the number of positive and negative eigenvalues. Expand
Spectral clustering with physical intuition on spring-mass dynamics
TLDR
This paper model a weighted graph of a data set as a mass–spring dynamical system, where a cluster is regarded as an oscillating single entity of aData set with similar properties, and describes how oscillation modes are related with eigenvectors of a graph Laplacian matrix of the data set. Expand
On Spectral Clustering: Analysis and an algorithm
TLDR
A simple spectral clustering algorithm that can be implemented using a few lines of Matlab is presented, and tools from matrix perturbation theory are used to analyze the algorithm, and give conditions under which it can be expected to do well. Expand
...
1
2
3
4
5
...