{"paper":{"title":"A Bayesian Adaptive Latent Mixture Model for Zero-Inflated Weighted Brain Connectome Analysis","license":"http://creativecommons.org/licenses/by/4.0/","headline":"A Bayesian mixture model represents each brain network as a simplex mixture of shared low-rank latent templates while separating edge presence from strength.","cross_cats":["stat.AP","stat.CO"],"primary_cat":"stat.ME","authors_text":"Hsin-Hsiung Huang, Teng Zhang, Yuh-Haur Chen","submitted_at":"2026-05-13T02:25:09Z","abstract_excerpt":"Replicated weighted networks often exhibit many structural zeros alongside heterogeneous non-zero edge strengths. In structural connectomics, this zero-inflation coincides with subjects expressing overlapping, rather than discrete, connectivity patterns. To address these features, we propose a Bayesian adaptive latent mixture model for zero-inflated weighted networks. Our approach represents each subject network as a simplex mixture of shared low-rank latent score matrices, integrated with a hurdle likelihood that separates edge existence from conditional edge strength. A sparsity-coupling par"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"The model recovers stable latent score patterns and heterogeneous subject-level mixtures in Human Connectome Project data; posterior consistency, local asymptotic normality, Bernstein-von Mises approximation, and predictive consistency hold for an identifiable quotient-space estimand under fixed-template scenario.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That subject networks are well-represented as simplex mixtures of a small number of shared low-rank latent score matrices, with the sparsity-coupling parameter correctly capturing dependence between absent edges and latent structure, and that template count selection via predictive fit yields an identifiable model.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"A Bayesian adaptive latent mixture model using simplex mixtures of low-rank latent score matrices and hurdle likelihoods for zero-inflated weighted brain connectomes, with posterior consistency and predictive consistency established.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"A Bayesian mixture model represents each brain network as a simplex mixture of shared low-rank latent templates while separating edge presence from strength.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"987e5beae32d5ec205a9112c7bc021bae5bc1d7d4c9174b216cbebab019c8043"},"source":{"id":"2605.12901","kind":"arxiv","version":1},"verdict":{"id":"bfcd51a8-5a52-420a-94b1-eec970cac1b9","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-14T18:53:26.257009Z","strongest_claim":"The model recovers stable latent score patterns and heterogeneous subject-level mixtures in Human Connectome Project data; posterior consistency, local asymptotic normality, Bernstein-von Mises approximation, and predictive consistency hold for an identifiable quotient-space estimand under fixed-template scenario.","one_line_summary":"A Bayesian adaptive latent mixture model using simplex mixtures of low-rank latent score matrices and hurdle likelihoods for zero-inflated weighted brain connectomes, with posterior consistency and predictive consistency established.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That subject networks are well-represented as simplex mixtures of a small number of shared low-rank latent score matrices, with the sparsity-coupling parameter correctly capturing dependence between absent edges and latent structure, and that template count selection via predictive fit yields an identifiable model.","pith_extraction_headline":"A Bayesian mixture model represents each brain network as a simplex mixture of shared low-rank latent templates while separating edge presence from strength."},"references":{"count":46,"sample":[{"doi":"","year":null,"title":"Journal of Machine Learning Research , year =","work_id":"ca28afaf-c19f-4888-b920-43fae2995da9","ref_index":1,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2007,"title":"Probabilistic diffusion tractography with multiple fibre orientations: What can we gain? , author=. NeuroImage , volume=. 2007 , publisher=","work_id":"152a855f-349a-478d-9055-b2e78136b5bc","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2017,"title":"Journal of the American Statistical Association , volume=","work_id":"4c03529f-9353-4073-9a95-04fb93315601","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2009,"title":"Accurate and robust brain image alignment using boundary-based registration , author=. NeuroImage , volume=. 2009 , publisher=","work_id":"99c460d8-ba8f-4dee-84da-eb8a138acb6d","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2024,"title":"Cai, Yuhua and Owen, Jonathan P. and Eriksson, M. and Reh, G. R. and Martin, L. and Irimia, A. and Davenport, N. D. and Mukherjee, P. and Mayer, A. R. , journal=. 2024 , doi=","work_id":"d8047f20-7077-43f1-96e8-4a352e4f49a2","ref_index":5,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":46,"snapshot_sha256":"69045fb31012f04d0f8633aea97d301d4147bb86d7e737475ce32970042ea274","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"}