{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:YDVSEMJPGCUHHTOZB7FPTVQM7G","short_pith_number":"pith:YDVSEMJP","schema_version":"1.0","canonical_sha256":"c0eb22312f30a873cdd90fcaf9d60cf9903a08d748852a2bfdf7d270fc8ffb13","source":{"kind":"arxiv","id":"2605.13620","version":1},"attestation_state":"computed","paper":{"title":"A Majorization-Minimization with Monte Carlo Approach for Hyperparameter Estimation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"M³C iterates converge with high probability to a critical point of the hyperparameter cost function by majorizing and Monte Carlo approximating the log-determinant.","cross_cats":["cs.NA"],"primary_cat":"math.NA","authors_text":"Arvind K. Saibaba, Elle Buser, Hugo D\\'iaz, Julianne Chung","submitted_at":"2026-05-13T14:50:12Z","abstract_excerpt":"We consider inverse problems with linear forward models and Gaussian priors, but with unknown hyperparameters that may arise from the model, the noise, or the specification of the prior. We model this using a hierarchical Bayes framework resulting in a posterior distribution that is non-Gaussian, in general, and challenging to sample from. Consequently, we use an empirical Bayes framework for estimating the maximum a posteriori estimate of the hyperpameters by considering the marginalized posterior distribution. However, the optimization problem is also computationally challenging due to the n"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":true,"formal_links_present":true},"canonical_record":{"source":{"id":"2605.13620","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"math.NA","submitted_at":"2026-05-13T14:50:12Z","cross_cats_sorted":["cs.NA"],"title_canon_sha256":"0b9a5587100759b577ef2fcc47bb18fc0ea32c256e0e56258bcebebb5ec724f1","abstract_canon_sha256":"d1ee412eeccaeafc21b97910ad9523d46e09e51eb00ae778a381fb3c54b59782"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T02:44:17.901862Z","signature_b64":"lOsliDIMTNV3ehKrA0PEKboENuTmsVMsnkCZRvzvw2tqOT8/Wh6/iE5OXRMnL1irp/t7d/yQZXrorAohhB8dCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c0eb22312f30a873cdd90fcaf9d60cf9903a08d748852a2bfdf7d270fc8ffb13","last_reissued_at":"2026-05-18T02:44:17.901364Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T02:44:17.901364Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"A Majorization-Minimization with Monte Carlo Approach for Hyperparameter Estimation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"M³C iterates converge with high probability to a critical point of the hyperparameter cost function by majorizing and Monte Carlo approximating the log-determinant.","cross_cats":["cs.NA"],"primary_cat":"math.NA","authors_text":"Arvind K. Saibaba, Elle Buser, Hugo D\\'iaz, Julianne Chung","submitted_at":"2026-05-13T14:50:12Z","abstract_excerpt":"We consider inverse problems with linear forward models and Gaussian priors, but with unknown hyperparameters that may arise from the model, the noise, or the specification of the prior. We model this using a hierarchical Bayes framework resulting in a posterior distribution that is non-Gaussian, in general, and challenging to sample from. Consequently, we use an empirical Bayes framework for estimating the maximum a posteriori estimate of the hyperpameters by considering the marginalized posterior distribution. However, the optimization problem is also computationally challenging due to the n"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"we propose a Majorization-Minimization with Monte Carlo approach, which we call M³C, for hyperparameter estimation. ... showing that under certain assumptions, the M³C iterates converge with high probability to a critical point of the original cost function.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"The convergence holds under certain assumptions on the validity of the majorization function and the accuracy of the Monte Carlo estimator for the log-determinant term.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"M³C replaces the hard hyperparameter optimization with a sequence of simpler problems using a majorant for the log-determinant approximated via Monte Carlo, with proven high-probability convergence to a critical point under assumptions.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"M³C iterates converge with high probability to a critical point of the hyperparameter cost function by majorizing and Monte Carlo approximating the log-determinant.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"80462877416dcf89adca7a210c3e6585fca2b15b684456a4cc5bb362e15ef36e"},"source":{"id":"2605.13620","kind":"arxiv","version":1},"verdict":{"id":"5a9b060f-7eee-42dd-bbb8-52b175b731fd","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-14T18:09:16.123101Z","strongest_claim":"we propose a Majorization-Minimization with Monte Carlo approach, which we call M³C, for hyperparameter estimation. ... showing that under certain assumptions, the M³C iterates converge with high probability to a critical point of the original cost function.","one_line_summary":"M³C replaces the hard hyperparameter optimization with a sequence of simpler problems using a majorant for the log-determinant approximated via Monte Carlo, with proven high-probability convergence to a critical point under assumptions.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"The convergence holds under certain assumptions on the validity of the majorization function and the accuracy of the Monte Carlo estimator for the log-determinant term.","pith_extraction_headline":"M³C iterates converge with high probability to a critical point of the hyperparameter cost function by majorizing and Monte Carlo approximating the log-determinant."},"references":{"count":189,"sample":[{"doi":"","year":null,"title":"Inverse Problems , volume=","work_id":"e30f37d1-0d7c-4e0b-bc0c-ef4526f13dab","ref_index":1,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":1989,"title":"The Gohberg Anniversary Collection: Volume I: The Calgary Conference and Matrix Theory Papers and Volume II: Topics in Analysis and Operator Theory , pages=","work_id":"4e85b78e-01df-4aa1-af01-1140ddcba2ce","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2017,"title":"Super-resolution imaging , author=. 2017 , publisher=","work_id":"880f19d3-b44b-4a8a-bee1-a2312de6d692","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2018,"title":"High-dimensional probability: An introduction with applications in data science , author=. 2018 , publisher=","work_id":"997454ae-f4d9-4a8a-b2c6-024b31656f08","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2026,"title":"Bayesian Statistical Methods: With Applications to Machine Learning , author=. 2026 , publisher=","work_id":"7b9bf3c8-e62d-46ac-8f3d-8fa87d13639e","ref_index":5,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":189,"snapshot_sha256":"a0c54bc4eca5fab70113629a48056939bb857f65804d041e39c07728f6b4a915","internal_anchors":1},"formal_canon":{"evidence_count":2,"snapshot_sha256":"41bf972ff9592e51ed4211762e536abe95022aee9fe4ce38812a1f00277ad0c7"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"2605.13620","created_at":"2026-05-18T02:44:17.901450+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.13620v1","created_at":"2026-05-18T02:44:17.901450+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.13620","created_at":"2026-05-18T02:44:17.901450+00:00"},{"alias_kind":"pith_short_12","alias_value":"YDVSEMJPGCUH","created_at":"2026-05-18T12:33:37.589309+00:00"},{"alias_kind":"pith_short_16","alias_value":"YDVSEMJPGCUHHTOZ","created_at":"2026-05-18T12:33:37.589309+00:00"},{"alias_kind":"pith_short_8","alias_value":"YDVSEMJP","created_at":"2026-05-18T12:33:37.589309+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":2,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/YDVSEMJPGCUHHTOZB7FPTVQM7G","json":"https://pith.science/pith/YDVSEMJPGCUHHTOZB7FPTVQM7G.json","graph_json":"https://pith.science/api/pith-number/YDVSEMJPGCUHHTOZB7FPTVQM7G/graph.json","events_json":"https://pith.science/api/pith-number/YDVSEMJPGCUHHTOZB7FPTVQM7G/events.json","paper":"https://pith.science/paper/YDVSEMJP"},"agent_actions":{"view_html":"https://pith.science/pith/YDVSEMJPGCUHHTOZB7FPTVQM7G","download_json":"https://pith.science/pith/YDVSEMJPGCUHHTOZB7FPTVQM7G.json","view_paper":"https://pith.science/paper/YDVSEMJP","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.13620&json=true","fetch_graph":"https://pith.science/api/pith-number/YDVSEMJPGCUHHTOZB7FPTVQM7G/graph.json","fetch_events":"https://pith.science/api/pith-number/YDVSEMJPGCUHHTOZB7FPTVQM7G/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/YDVSEMJPGCUHHTOZB7FPTVQM7G/action/timestamp_anchor","attest_storage":"https://pith.science/pith/YDVSEMJPGCUHHTOZB7FPTVQM7G/action/storage_attestation","attest_author":"https://pith.science/pith/YDVSEMJPGCUHHTOZB7FPTVQM7G/action/author_attestation","sign_citation":"https://pith.science/pith/YDVSEMJPGCUHHTOZB7FPTVQM7G/action/citation_signature","submit_replication":"https://pith.science/pith/YDVSEMJPGCUHHTOZB7FPTVQM7G/action/replication_record"}},"created_at":"2026-05-18T02:44:17.901450+00:00","updated_at":"2026-05-18T02:44:17.901450+00:00"}