Robust Identification of Graph Structure

Partial correlations (PCs) and the related inverse covariance matrix adopted by graphical lasso, are widely applicable tools for learning graph connectivity given nodal observations. The resultant estimators however, can be sensitive to outliers. Robust approaches developed so far to cope with outliers do not (explicitly) account for nonlinear interactions possibly present among nodal processes. This can hurt the identification of graph connectivity, merely due to model mismatch. To overcome this limitation, a novel formulation of robust PC is introduced based on nonlinear kernel functions. The proposed scheme leverages robust ridge regression techniques, spectral Fourier feature based kernel approximants, and robust association measures. Numerical tests on synthetic and real data illustrate the potential of the novel approach.

• Publication year

  2023

• Venue

  IEEE International Workshop on Computational Advances in Multi-Sensor Adaptive Processing

• Publication date

  2023-12-10

• Fields of study

  Mathematics, Computer Science

• Identifiers
  DOI 10.1109/CAMSAP58249.2023.10403439
• External record

  Open on Semantic Scholar

• Source metadata

  Semantic Scholar

• No claims are published for this paper.

• No concepts are published for this paper.

• Connecting the Dots: Identifying Network Structure via Graph Signal Processing

  2018cited by this paper
• Robust Statistics for Signal Processing

  2018cited by this paper
• Topology Identification and Learning over Graphs: Accounting for Nonlinearities and Dynamics

  2018cited by this paper
• Learning Graphs From Data: A Signal Representation Perspective

  2018cited by this paper
• Multi-kernel based nonlinear models for connectivity identification of brain networks

  2016cited by this paper
• Universal resilience patterns in complex networks

  2016cited by this paper
• The influence function of penalized regression estimators

  2015cited by this paper
• Robust estimation of precision matrices under cellwise contamination

  2015cited by this paper
• Robust sparse Gaussian graphical modeling

  2015cited by this paper
• Robust and sparse estimators for linear regression models

  2015cited by this paper
• Robust Ridge Regression for High-Dimensional Data

  2011influential reference
• Robust Estimation of the Correlation Coefficient: An Attempt of Survey

  2011cited by this paper
• Sparse Spectrum Gaussian Process Regression

  2010cited by this paper
• Realising the future: forecasting with high frequency based volatility (HEAVY) models

  2010cited by this paper
• Robust graphical modeling of gene networks using classical and alternative t-distributions

  2010cited by this paper
• Statistical Analysis of Network Data

  2009influential reference
• Random Features for Large-Scale Kernel Machines

  2007cited by this paper
• Emergence of scaling in random networks

  1999cited by this paper
• Robust estimation in structured linear regression

  1996influential reference
• A robust partial correlation measure

  1995cited by this paper
• Alternatives to the Median Absolute Deviation

  1993cited by this paper
• On the optimality of S-estimators☆

  1992cited by this paper
• HIGH BREAKDOWN-POINT AND HIGH EFFICIENCY ROBUST ESTIMATES FOR REGRESSION

  1987cited by this paper
• On robust estimation of the location parameter

  1980cited by this paper
• ROBUST ESTIMATES, RESIDUALS, AND OUTLIER DETECTION WITH MULTIRESPONSE DATA

  1972cited by this paper
• Agreement

  1931cited by this paper

• No citing papers are available for this paper.