arXiv subjects

Carey E. Priebe

Publications and source records attributed to Carey E. Priebe.

At least 163 records · Page 9Linked to original sources

arXiv2014

Seeded graph matching for correlated Erdős-Rényi graphs

Graph matching is an important problem in machine learning and pattern recognition. Herein, we present theoretical and practical results on the consistency of graph matching for estimating a latent alignment function between the vertex sets of two graphs, as well as subsequent algorithmic implications when the latent alignment is partially observed. In the correlated Erdős-Rényi graph setting, we prove that graph matching provides a strongly consistent estimate of the latent alignment in the presence of even modest correlation. We then investigate a tractable, restricted-focus version of graph matching, which is only concerned with adjacency involving vertices in a partial observation of the latent alignment; we prove that a logarithmic number of vertices whose alignment is known is sufficient for this restricted-focus version of graph matching to yield a strongly consistent estimate of the latent alignment of the remaining vertices. We show how Frank-Wolfe methodology for approximate graph matching, when there is a partially observed latent alignment, inherently incorporates this restricted focus graph matching. Lastly, we illustrate the relationship between seeded graph matching and restricted-focus graph matching by means of an illuminating example from human connectomics.

math.OC↗

arXiv2014

Statistical inference on errorfully observed graphs

Statistical inference on graphs is a burgeoning field in the applied and theoretical statistics communities, as well as throughout the wider world of science, engineering, business, etc. In many applications, we are faced with the reality of errorfully observed graphs. That is, the existence of an edge between two vertices is based on some imperfect assessment. In this paper, we consider a graph $G = (V,E)$. We wish to perform an inference task -- the inference task considered here is "vertex classification". However, we do not observe $G$; rather, for each potential edge $uv \in {{V}\choose{2}}$ we observe an "edge-feature" which we use to classify $uv$ as edge/not-edge. Thus we errorfully observe $G$ when we observe the graph $\widetilde{G} = (V,\widetilde{E})$ as the edges in $\widetilde{E}$ arise from the classifications of the "edge-features", and are expected to be errorful. Moreover, we face a quantity/quality trade-off regarding the edge-features we observe -- more informative edge-features are more expensive, and hence the number of potential edges that can be assessed decreases with the quality of the edge-features. We studied this problem by formulating a quantity/quality tradeoff for a simple class of random graphs model, namely the stochastic blockmodel. We then consider a simple but optimal vertex classifier for classifying $v$ and we derive the optimal quantity/quality operating point for subsequent graph inference in the face of this trade-off. The optimal operating points for the quantity/quality trade-off are surprising and illustrate the issue that methods for intermediate tasks should be chosen to maximize performance for the ultimate inference task. Finally, we investigate the quantity/quality tradeoff for errorful obesrvations of the {\it C.\ elegans} connectome graph.

stat.ML↗

arXiv2014

Generalized Canonical Correlation Analysis for Classification

For multiple multivariate data sets, we derive conditions under which Generalized Canonical Correlation Analysis (GCCA) improves classification performance of the projected datasets, compared to standard Canonical Correlation Analysis (CCA) using only two data sets. We illustrate our theoretical results with simulations and a real data experiment.

stat.ML↗

arXiv2013

A central limit theorem for scaled eigenvectors of random dot product graphs

We prove a central limit theorem for the components of the largest eigenvectors of the adjacency matrix of a finite-dimensional random dot product graph whose true latent positions are unknown. In particular, we follow the methodology outlined in \citet{sussman2012universally} to construct consistent estimates for the latent positions, and we show that the appropriately scaled differences between the estimated and true latent positions converge to a mixture of Gaussian random variables. As a corollary, we obtain a central limit theorem for the first eigenvector of the adjacency matrix of an Erdös-Renyi random graph.

math.ST↗

arXiv2013

Computing Scalable Multivariate Glocal Invariants of Large (Brain-) Graphs

Graphs are quickly emerging as a leading abstraction for the representation of data. One important application domain originates from an emerging discipline called "connectomics". Connectomics studies the brain as a graph; vertices correspond to neurons (or collections thereof) and edges correspond to structural or functional connections between them. To explore the variability of connectomes---to address both basic science questions regarding the structure of the brain, and medical health questions about psychiatry and neurology---one can study the topological properties of these brain-graphs. We define multivariate glocal graph invariants: these are features of the graph that capture various local and global topological properties of the graphs. We show that the collection of features can collectively be computed via a combination of daisy-chaining, sparse matrix representation and computations, and efficient approximations. Our custom open-source Python package serves as a back-end to a Web-service that we have created to enable researchers to upload graphs, and download the corresponding invariants in a number of different formats. Moreover, we built this package to support distributed processing on multicore machines. This is therefore an enabling technology for network science, lowering the barrier of entry by providing tools to biologists and analysts who otherwise lack these capabilities. As a demonstration, we run our code on 120 brain-graphs, each with approximately 16M vertices and up to 90M edges.

eess.SY↗

arXiv2013

Detecting Time-dependent Structure in Network Data via a New Class of Latent Process Models

We introduce a new class of latent process models for dynamic relational network data with the goal of detecting time-dependent structure. Network data are often observed over time, and static network models for such data may fail to capture relevant dynamic features. We present a new technique for identifying the emergence or disappearance of distinct subpopulations of vertices. In this formulation, a network is observed over time, with attributed edges appearing at random times. At unknown time points, subgroups of vertices may exhibit a change in behavior. Such changes may take the form of a change in the overall probability of connection within or between subgroups, or a change in the distribution of edge attributes. A mixture distribution for latent vertex positions is used to detect heterogeneities in connectivity behavior over time and over vertices. The probability of edges with various attributes at a given time is modeled using a latent-space stochastic process associated with each vertex. A random dot product model is used to describe the dependency structure of the graph. As an application we analyze the Enron email corpus.

stat.ME↗

arXiv2013

Locality statistics for anomaly detection in time series of graphs

The ability to detect change-points in a dynamic network or a time series of graphs is an increasingly important task in many applications of the emerging discipline of graph signal processing. This paper formulates change-point detection as a hypothesis testing problem in terms of a generative latent position model, focusing on the special case of the Stochastic Block Model time series. We analyze two classes of scan statistics, based on distinct underlying locality statistics presented in the literature. Our main contribution is the derivation of the limiting distributions and power characteristics of the competing scan statistics. Performance is compared theoretically, on synthetic data, and on the Enron email corpus. We demonstrate that both statistics are admissible in one simple setting, while one of the statistics is inadmissible a second setting.

stat.AP↗

arXiv2013

Universally consistent vertex classification for latent positions graphs

In this work we show that, using the eigen-decomposition of the adjacency matrix, we can consistently estimate feature maps for latent position graphs with positive definite link function $κ$, provided that the latent positions are i.i.d. from some distribution F. We then consider the exploitation task of vertex classification where the link function $κ$ belongs to the class of universal kernels and class labels are observed for a number of vertices tending to infinity and that the remaining vertices are to be classified. We show that minimization of the empirical $φ$-risk for some convex surrogate $φ$ of 0-1 loss over a class of linear classifiers with increasing complexities yields a universally consistent classifier, that is, a classification rule with error converging to Bayes optimal for any distribution F.

stat.ML↗

arXiv2013

Out-of-sample Extension for Latent Position Graphs

We consider the problem of vertex classification for graphs constructed from the latent position model. It was shown previously that the approach of embedding the graphs into some Euclidean space followed by classification in that space can yields a universally consistent vertex classifier. However, a major technical difficulty of the approach arises when classifying unlabeled out-of-sample vertices without including them in the embedding stage. In this paper, we studied the out-of-sample extension for the graph embedding step and its impact on the subsequent inference tasks. We show that, under the latent position graph model and for sufficiently large $n$, the mapping of the out-of-sample vertices is close to its true latent position. We then demonstrate that successful inference for the out-of-sample vertices is possible.

stat.ML↗

arXiv2013

Enhanced Archaeological Predictive Modelling in Space Archaeology

Identifying and preserving archaeological sites before they are destroyed is a very important issue. In this paper, we develop a greatly improved archaeological predictive model $APM_{enhanced}$ that predicts where archaeological sites will be found. This approach is applied to remotely-sensed multispectral bands and a single topographical band obtained from advanced remote sensing technologies such as satellites and Airborne Laser Scanning (ALS). Our $APM_{enhanced}$ is composed of band transformation, image analysis, feature extraction and classification. We evaluate our methodology on the sensor bands over Ft.Irwin, CA, USA. A nested bi-loop cross-validation and receiver operating characteristics curves are used to assess the performance of our algorithm. We first validate our method on the east swath of Ft.Irwin and then test on a separate dataset from the west swath of Ft.Irwin. A convex combination of two methodologies: $APM_{conventional}$, which has been used among archaeologists for many years, and our $APM_{enhanced}$, is demonstrated to yield superior classification performance compared to either alone at low false negative rates. We compare the performance of our methodology on different band combinations, chosen based on the archaeological importance for these sensor bands. We also compare the two types of $APM$s in the aspects of input data, output values, practicality and transferability.

stat.AP↗

arXiv2012

Anomaly Detection in Time Series of Graphs using Fusion of Graph Invariants

Given a time series of graphs G(t) = (V, E(t)), t = 1, 2, ..., where the fixed vertex set V represents "actors" and an edge between vertex u and vertex v at time t (uv \in E(t)) represents the existence of a communications event between actors u and v during the tth time period, we wish to detect anomalies and/or change points. We consider a collection of graph features, or invariants, and demonstrate that adaptive fusion provides superior inferential efficacy compared to naive equal weighting for a certain class of anomaly detection problems. Simulation results using a latent process model for time series of graphs, as well as illustrative experimental results for a time series of graphs derived from the Enron email data, show that a fusion statistic can provide superior inference compared to individual invariants alone. These results also demonstrate that an adaptive weighting scheme for fusion of invariants performs better than naive equal weighting.

stat.ML↗

arXiv2012

Shuffled Graph Classification: Theory and Connectome Applications

We develop a formalism to address statistical pattern recognition of graph valued data. Of particular interest is the case of all graphs having the same number of uniquely labeled vertices. When the vertex labels are latent, such graphs are called shuffled graphs. Our formalism provides insight to trivially answer a number of open statistical questions including: (i) under what conditions does shuffling the vertices degrade classification performance and (ii) do universally consistent graph classifiers exist? The answers to these questions lead to practical heuristic algorithms with state-of-the-art finite sample performance, in agreement with our theoretical asymptotics.

q-bio.QM↗

arXiv2012

Generalized Canonical Correlation Analysis for Disparate Data Fusion

Manifold matching works to identify embeddings of multiple disparate data spaces into the same low-dimensional space, where joint inference can be pursued. It is an enabling methodology for fusion and inference from multiple and massive disparate data sources. In this paper we focus on a method called Canonical Correlation Analysis (CCA) and its generalization Generalized Canonical Correlation Analysis (GCCA), which belong to the more general Reduced Rank Regression (RRR) framework. We present an efficiency investigation of CCA and GCCA under different training conditions for a particular text document classification task.

stat.ML↗

arXiv2012

Consistent adjacency-spectral partitioning for the stochastic block model when the model parameters are unknown

For random graphs distributed according to a stochastic block model, we consider the inferential task of partioning vertices into blocks using spectral techniques. Spectral partioning using the normalized Laplacian and the adjacency matrix have both been shown to be consistent as the number of vertices tend to infinity. Importantly, both procedures require that the number of blocks and the rank of the communication probability matrix are known, even as the rest of the parameters may be unknown. In this article, we prove that the (suitably modified) adjacency-spectral partitioning procedure, requiring only an upper bound on the rank of the communication probability matrix, is consistent. Indeed, this result demonstrates a robustness to model mis-specification; an overestimate of the rank may impose a moderate performance penalty, but the procedure is still consistent. Furthermore, we extend this procedure to the setting where adjacencies may have multiple modalities and we allow for either directed or undirected graphs.

stat.ME↗

arXiv2012

Universally Consistent Latent Position Estimation and Vertex Classification for Random Dot Product Graphs

In this work we show that, using the eigen-decomposition of the adjacency matrix, we can consistently estimate latent positions for random dot product graphs provided the latent positions are i.i.d. from some distribution. If class labels are observed for a number of vertices tending to infinity, then we show that the remaining vertices can be classified with error converging to Bayes optimal using the $k$-nearest-neighbors classification rule. We evaluate the proposed methods on simulated data and a graph derived from Wikipedia.

stat.ML↗

arXiv2012

Graph Classification using Signal-Subgraphs: Applications in Statistical Connectomics

This manuscript considers the following "graph classification" question: given a collection of graphs and associated classes, how can one predict the class of a newly observed graph? To address this question we propose a statistical model for graph/class pairs. This model naturally leads to a set of estimators to identify the class-conditional signal, or "signal-subgraph," defined as the collection of edges that are probabilistically different between the classes. The estimators admit classifiers which are asymptotically optimal and efficient, but differ by their assumption about the "coherency" of the signal-subgraph (coherency is the extent to which the signal-edges "stick together" around a common subset of vertices). Via simulation, the best estimator is shown to be not just a function of the coherency of the model, but also the number of training samples. These estimators are employed to address a contemporary neuroscience question: can we classify "connectomes" (brain-graphs) according to sex? The answer is yes, and significantly better than all benchmark algorithms considered. Synthetic data analysis demonstrates that even when the model is correct, given the relatively small number of training samples, the estimated signal-subgraph should be taken with a grain of salt. We conclude by discussing several possible extensions.

stat.AP↗

arXiv2012

Bayesian Vertex Nomination

Consider an attributed graph whose vertices are colored green or red, but only a few are observed to be red. The color of the other vertices is unobserved. Typically, the unknown total number of red vertices is small. The vertex nomination problem is to nominate one of the unobserved vertices as being red. The edge set of the graph is a subset of the set of unordered pairs of vertices. Suppose that each edge is also colored green or red and this is observed for all edges. The context statistic of a vertex is defined as the number of observed red vertices connected to it, and its content statistic is the number of red edges incident to it. Assuming that these statistics are independent between vertices and that red edges are more likely between red vertices, Coppersmith and Priebe (2012) proposed a likelihood model based on these statistics. Here, we formulate a Bayesian model using the proposed likelihood together with prior distributions chosen for the unknown parameters and unobserved vertex colors. From the resulting posterior distribution, the nominated vertex is the one with the highest posterior probability of being red. Inference is conducted using a Metropolis-within-Gibbs algorithm, and performance is illustrated by a simulation study. Results show that (i) the Bayesian model performs significantly better than chance; (ii) the probability of correct nomination increases with increasing posterior probability that the nominated vertex is red; and (iii) the Bayesian model either matches or performs better than the method in Coppersmith and Priebe. An application example is provided using the Enron email corpus, where vertices represent Enron employees and their associates, observed red vertices are known fraudsters, red edges represent email communications perceived as fraudulent, and we wish to identify one of the latent vertices as most likely to be a fraudster.

stat.ME↗

arXiv2012

A consistent adjacency spectral embedding for stochastic blockmodel graphs

We present a method to estimate block membership of nodes in a random graph generated by a stochastic blockmodel. We use an embedding procedure motivated by the random dot product graph model, a particular example of the latent position model. The embedding associates each node with a vector; these vectors are clustered via minimization of a square error criterion. We prove that this method is consistent for assigning nodes to blocks, as only a negligible number of nodes will be mis-assigned. We prove consistency of the method for directed and undirected graphs. The consistent block assignment makes possible consistent parameter estimation for a stochastic blockmodel. We extend the result in the setting where the number of blocks grows slowly with the number of nodes. Our method is also computationally feasible even for very large graphs. We compare our method to Laplacian spectral clustering through analysis of simulated data and a graph derived from Wikipedia documents.

stat.ML↗