Maximum likelihood estimation of sparse networks with missing observations
From MaRDI portal
Publication:2242863
Abstract: Estimating the matrix of connections probabilities is one of the key questions when studying sparse networks. In this work, we consider networks generated under the sparse graphon model and the in-homogeneous random graph model with missing observations. Using the Stochastic Block Model as a parametric proxy, we bound the risk of the maximum likelihood estimator of network connections probabilities , and show that it is minimax optimal. When risk is measured in Frobenius norm, no estimator running in polynomial time has been shown to attain the minimax optimal rate of convergence for this problem. Thus, maximum likelihood estimation is of particular interest as computationally efficient approximations to it have been proposed in the literature and are often used in practice.
Recommendations
- Consistency and asymptotic normality of stochastic block models estimators from sampled data
- Oracle inequalities for network models and sparse graphon estimation
- Estimating sparse networks with hubs
- Maximum likelihood estimation in the -model
- Asymptotic normality of maximum likelihood and its variational approximation for stochastic blockmodels
Cites work
- 1-bit matrix completion
- A goodness-of-fit test for stochastic block models
- A nonparametric view of network models and Newman–Girvan and other modularities
- Achieving Exact Cluster Recovery Threshold via Semidefinite Programming: Extensions
- Adaptive multinomial matrix completion
- Asymptotic normality of maximum likelihood and its variational approximation for stochastic blockmodels
- Community detection thresholds and the weak Ramanujan property
- Consistency and asymptotic normality of stochastic block models estimators from sampled data
- Consistency of maximum-likelihood and variational estimators in the stochastic block model
- Detection of structurally homogeneous subsets in graphs
- Estimating network edge probabilities by neighbourhood smoothing
- Estimation of (near) low-rank matrices with noise and high-dimensional scaling
- Large networks and graph limits
- Likelihood-based model selection for stochastic block models
- Mathematical Foundations of Infinite-Dimensional Statistical Models
- Matrix estimation by universal singular value thresholding
- Modeling heterogeneity in random graphs through latent space models: a selective review
- Modeling social networks from sampled data
- Network cross-validation for determining the number of communities in network data
- Noisy low-rank matrix completion with general sampling distribution
- Nonbacktracking spectrum of random graphs: community detection and nonregular Ramanujan graphs
- Nuclear-norm penalization and optimal rates for noisy low-rank matrix completion
- On semidefinite relaxations for the block model
- Optimal estimation and completion of matrices with biclustering structures
- Oracle inequalities for network models and sparse graphon estimation
- Overlapping stochastic block models with application to the French political blogosphere
- Partial recovery bounds for clustering with the relaxed K-means
- Pseudo-likelihood methods for community detection in large sparse networks
- Rate-optimal graphon estimation
- Spectral clustering and the high-dimensional stochastic blockmodel
- Statistical mechanics of complex networks
- Variational Inference for Stochastic Block Models From Sampled Data
Cited in
(10)- Consistency of the maximum likelihood and variational estimators in a dynamic stochastic block model
- Sparse estimation in linear dynamic networks using the stable spline horseshoe prior
- Consistency and asymptotic normality of stochastic block models estimators from sampled data
- Estimation of Sparse Directional Connectivity With Expectation Maximization
- Jackknife empirical likelihood: small bandwidth, sparse network and high-dimensional asymptotics
- Oracle inequalities for network models and sparse graphon estimation
- Maximum likelihood estimation based on the Laplace approximation for p2 network regression models
- Conformal link prediction for false discovery rate control
- Denoising over networks with applications to partially observed epidemics
- Iterative maximum likelihood on networks
This page was built for publication: Maximum likelihood estimation of sparse networks with missing observations
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2242863)
