eduzhai > Applied Sciences > Engineering >

Adaptive support driven Bayesian reweighted algorithm for sparse signal recovery

  • king
  • (0) Download
  • 20210506
  • Save

... pages left unread,continue reading

Document pages: 7 pages

Abstract: Sparse learning has been widely studied to capture critical information fromenormous data sources in the filed of system identification. Often, it isessential to understand internal working mechanisms of unknown systems (e.g.biological networks) in addition to input-output relationships. For thispurpose, various feature selection techniques have been developed. For example,sparse Bayesian learning (SBL) was proposed to learn major features from adictionary of basis functions, which makes identified models interpretable.Reweighted L1-regularization algorithms are often applied in SBL to solveoptimization problems. However, they are expensive in both computation andmemory aspects, thus not suitable for large-scale problems. This paper proposesan adaptive support driven Bayesian reweighted (ASDBR) algorithm for sparsesignal recovery. A restart strategy based on shrinkage-thresholding isdeveloped to conduct adaptive support estimate, which can effectively reducecomputation burden and memory demands. Moreover, ASDBR accurately extractsmajor features and excludes redundant information from large datasets.Numerical experiments demonstrate the proposed algorithm outperformsstate-of-the-art methods.

Please select stars to rate!

         

0 comments Sign in to leave a comment.

    Data loading, please wait...
×