• Corpus ID: 3344994

Optimal Single Sample Tests for Structured versus Unstructured Network Data

@inproceedings{Bresler2018OptimalSS,
  title={Optimal Single Sample Tests for Structured versus Unstructured Network Data},
  author={Guy Bresler and Dheeraj M. Nagaraj},
  booktitle={COLT},
  year={2018}
}
We study the problem of testing, using only a single sample, between mean field distributions (like Curie-Weiss, Erd\H{o}s-R\'enyi) and structured Gibbs distributions (like Ising model on sparse graphs and Exponential Random Graphs). Our goal is to test without knowing the parameter values of the underlying models: only the \emph{structure} of dependencies is known. We develop a new approach that applies to both the Ising and Exponential Random Graph settings based on a general and natural… 
Goodness-of-Fit Tests for Inhomogeneous Random Graphs
TLDR
This paper derives the optimal minimax sample complexities for the goodness-of-fit problem in various natural norms, such as the Frobenius norm and the operator norm, and proposes practical implementations of natural test statistics, using their asymptotic distributions and through the parametric bootstrap.
Limits on Testing Structural Changes in Ising Models
TLDR
Novel information-theoretic limits on detecting sparse changes in Ising models are presented, suggesting that testing of graphical models may not be amenable to concepts such as restricted strong convexity leveraged for sparsity pattern recovery, and algorithm development instead should be directed towards detection of large changes.
Logistic-Regression with peer-group effects via inference in higher order Ising models
TLDR
This work model binary outcomes on a network as a higher-order spin glass, where the behavior of an individual depends on a linear function of their own vector of covariates and some polynomial function of others, capturing peer-group effects.
Statistical Estimation from Dependent Data
We consider a general statistical estimation problem wherein binary labels across different observations are not independent conditioned on their feature vectors, but dependent, capturing settings
Phase Transitions for Detecting Latent Geometry in Random Graphs
TLDR
It is proved that the random intersection graph converges in total variation to G(n, p) when d = \tilde{\omega}(n^3) and does not if $d = o(n*3)$, resolving an open problem in Fill et al. (2018).
A Stein Goodness of fit Test for Exponential Random Graph Models
We propose and analyse a novel nonparametric goodness-of-fit testing procedure for exchangeable exponential random graph models (ERGMs) when a single network realisation is observed. The test
Modern challenges in distribution testing
TLDR
The goal of this dissertation is to identify and address several contemporary challenges in distribution testing and make progress in answering the following questions.
High Dimensional Logistic Regression Under Network Dependence
TLDR
A penalized maximum pseudo-likelihood method for estimating the network peer-effect and the effect of the covariates (the regression coefficients), which, in addition to handling the high-dimensionality of the parameters, conveniently avoids the computational intractability of the maximum likelihood approach.
Stein’s method for stationary distributions of Markov chains and application to Ising models
We develop a new technique, based on Stein's method, for comparing two stationary distributions of irreducible Markov Chains whose update rules are `close enough'. We apply this technique to compare
STEIN’S METHOD FOR STATIONARY DISTRIBUTIONS OF MARKOV CHAINS AND APPLICATION TO ISING MODELS1
We develop a new technique, based on Stein’s method, for comparing two stationary distributions of irreducible Markov chains whose update rules are close in a certain sense. We apply this technique
...
...

References

SHOWING 1-10 OF 31 REFERENCES
Two-Sample Tests for Large Random Graphs Using Network Statistics
TLDR
The main contribution of the paper is a general formulation of the problem based on concentration of network statistics, and consequently, a consistent two-sample test that arises as the natural solution for this problem.
Testing for high‐dimensional geometry in random graphs
TLDR
The proof of the detection lower bound is based on a new bound on the total variation distance between a Wishart matrix and an appropriately normalized GOE matrix and a conjecture for the optimal detection boundary is made.
Testing Bayesian Networks
TLDR
This work initiates a systematic investigation of testing high-dimensional structured distributions by focusing on testing Bayesian networks – the prototypical family of directed graphical models, and develops the first non-trivial efficient testing algorithms for these problems and corresponding information-theoretic lower bounds.
Testing Network Structure Using Relations Between Small Subgraph Probabilities
TLDR
The results show how global structural characteristics of networks can be inferred from local subgraph frequencies, without requiring the global community structure to be explicitly estimated.
A Statistical Model for Motifs Detection
TLDR
A statistical model for the problem of finding subgraphs with specified topology in an otherwise random graph and two polynomial-time algorithms for solving the same problem: a spectral algorithm and a semidefinite programming (SDP) relaxation.
Exponential random graphs behave like mixtures of stochastic block models
TLDR
This work generalizes some of the results in a paper by Chatterjee and Diaconis from the dense regime to the sparse regime and strengthens their bounds from the cut-metrics to the one-metric.
Stein’s method for stationary distributions of Markov chains and application to Ising models
We develop a new technique, based on Stein's method, for comparing two stationary distributions of irreducible Markov Chains whose update rules are `close enough'. We apply this technique to compare
Universality of the mean-field for the Potts model
We consider the Potts model with q colors on a sequence of weighted graphs with adjacency matrices $$A_n$$An, allowing for both positive and negative weights. Under a mild regularity condition on
Concentration inequalities for polynomials of contracting Ising models
We study the concentration of a degree-$d$ polynomial of the $N$ spins of a general Ising model, in the regime where single-site Glauber dynamics is contracting. For $d=1$, Gaussian concentration was
Concentration Inequalities - A Nonasymptotic Theory of Independence
TLDR
Deep connections with isoperimetric problems are revealed whilst special attention is paid to applications to the supremum of empirical processes.
...
...