2016/03/23 by Payal Tyagi, Alessia Marruzzo, Andrea Pagnani +3 · 10 citations
Mathematics · Physics and Astronomy · #Algorithm #Artificial intelligence #Computer science #Decimation #Inference #Inverse problem #Mathematical analysis #Mathematics #Monte Carlo method #Random lasers and scattering media #Regularization (linguistics) #Spectroscopy and Quantum Chemical Studies #Statistical inference #Statistics #Theoretical and Computational Physics #cond-mat.dis-nn #cond-mat.stat-mech #physics.optics
paper · pdf · doi:10.1103/physrevb.94.024203
published in Physical review. B./Physical review. B 94(2) (American Physical Society) · 10 pages, 12 figures
arxiv created 2016/03/23 · openalex created_date 2016/06/24 · openalex publication_date 2016/07/15 · arxiv updated 2016/07/20 · openalex updated_date 2026/08/05
We implement a pseudolikelihood approach with l1 and l2 regularizations as well as the recently introduced pseudolikelihood with decimation procedure to the inverse problem in continuous spin models on arbitrary networks, with arbitrarily disordered couplings. Performances of the approaches are tested against data produced by Monte Carlo numerical simulations and compared also to previously studied fully connected mean-field-based inference techniques. The results clearly show that the best network reconstruction is obtained through the decimation scheme, which also allows us to make the inference down to lower temperature regimes. Possible applications to phasor models for light propagation in random media are proposed and discussed.