{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:B3XVLFOHVLBZD3UL3BL6FES6IK","short_pith_number":"pith:B3XVLFOH","schema_version":"1.0","canonical_sha256":"0eef5595c7aac391ee8bd857e2925e42a0678aa71cf54ed7b6b8ba09fc4505ab","source":{"kind":"arxiv","id":"2406.17748","version":1},"attestation_state":"computed","paper":{"title":"A New Perspective on Shampoo's Preconditioner","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["math.OC","stat.ML"],"primary_cat":"cs.LG","authors_text":"Depen Morwani, Eran Malach, Itai Shapira, Lucas Janson, Nikhil Vyas, Sham Kakade","submitted_at":"2024-06-25T17:34:51Z","abstract_excerpt":"Shampoo, a second-order optimization algorithm which uses a Kronecker product preconditioner, has recently garnered increasing attention from the machine learning community. The preconditioner used by Shampoo can be viewed either as an approximation of the Gauss--Newton component of the Hessian or the covariance matrix of the gradients maintained by Adagrad. We provide an explicit and novel connection between the $\\textit{optimal}$ Kronecker product approximation of these matrices and the approximation made by Shampoo. Our connection highlights a subtle but common misconception about Shampoo's"},"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":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2406.17748","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-06-25T17:34:51Z","cross_cats_sorted":["math.OC","stat.ML"],"title_canon_sha256":"72f2007d57a19c9b262655bfcf6ebfadeeb8c127d45f5b3f2bb034cd7a1a6ed1","abstract_canon_sha256":"eae43c3b17984db9064d2b597d859519f2c990daee0929643e00930989a3d822"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:36:39.452942Z","signature_b64":"dcgXmkZiwAryj5e+/IcwxmxRaK3HEkRxKT3Q3U/nYoPXLL67joXvAPs0O38Hqas8BNyxTCfiIahNP2qBv3bsBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0eef5595c7aac391ee8bd857e2925e42a0678aa71cf54ed7b6b8ba09fc4505ab","last_reissued_at":"2026-07-05T08:36:39.452445Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:36:39.452445Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"A New Perspective on Shampoo's Preconditioner","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["math.OC","stat.ML"],"primary_cat":"cs.LG","authors_text":"Depen Morwani, Eran Malach, Itai Shapira, Lucas Janson, Nikhil Vyas, Sham Kakade","submitted_at":"2024-06-25T17:34:51Z","abstract_excerpt":"Shampoo, a second-order optimization algorithm which uses a Kronecker product preconditioner, has recently garnered increasing attention from the machine learning community. The preconditioner used by Shampoo can be viewed either as an approximation of the Gauss--Newton component of the Hessian or the covariance matrix of the gradients maintained by Adagrad. We provide an explicit and novel connection between the $\\textit{optimal}$ Kronecker product approximation of these matrices and the approximation made by Shampoo. Our connection highlights a subtle but common misconception about Shampoo's"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.17748","kind":"arxiv","version":1},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2406.17748/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","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"},"aliases":[{"alias_kind":"arxiv","alias_value":"2406.17748","created_at":"2026-07-05T08:36:39.452503+00:00"},{"alias_kind":"arxiv_version","alias_value":"2406.17748v1","created_at":"2026-07-05T08:36:39.452503+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.17748","created_at":"2026-07-05T08:36:39.452503+00:00"},{"alias_kind":"pith_short_12","alias_value":"B3XVLFOHVLBZ","created_at":"2026-07-05T08:36:39.452503+00:00"},{"alias_kind":"pith_short_16","alias_value":"B3XVLFOHVLBZD3UL","created_at":"2026-07-05T08:36:39.452503+00:00"},{"alias_kind":"pith_short_8","alias_value":"B3XVLFOH","created_at":"2026-07-05T08:36:39.452503+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":4,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.04662","citing_title":"Why Muon Outperforms Adam: A Curvature Perspective","ref_index":85,"is_internal_anchor":false},{"citing_arxiv_id":"2505.23737","citing_title":"On the Convergence Analysis of Muon","ref_index":16,"is_internal_anchor":false},{"citing_arxiv_id":"2409.20325","citing_title":"Old Optimizer, New Norm: An Anthology","ref_index":30,"is_internal_anchor":false},{"citing_arxiv_id":"2605.06316","citing_title":"Pro-KLShampoo: Projected KL-Shampoo with Whitening Recovered by Orthogonalization","ref_index":13,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/B3XVLFOHVLBZD3UL3BL6FES6IK","json":"https://pith.science/pith/B3XVLFOHVLBZD3UL3BL6FES6IK.json","graph_json":"https://pith.science/api/pith-number/B3XVLFOHVLBZD3UL3BL6FES6IK/graph.json","events_json":"https://pith.science/api/pith-number/B3XVLFOHVLBZD3UL3BL6FES6IK/events.json","paper":"https://pith.science/paper/B3XVLFOH"},"agent_actions":{"view_html":"https://pith.science/pith/B3XVLFOHVLBZD3UL3BL6FES6IK","download_json":"https://pith.science/pith/B3XVLFOHVLBZD3UL3BL6FES6IK.json","view_paper":"https://pith.science/paper/B3XVLFOH","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2406.17748&json=true","fetch_graph":"https://pith.science/api/pith-number/B3XVLFOHVLBZD3UL3BL6FES6IK/graph.json","fetch_events":"https://pith.science/api/pith-number/B3XVLFOHVLBZD3UL3BL6FES6IK/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/B3XVLFOHVLBZD3UL3BL6FES6IK/action/timestamp_anchor","attest_storage":"https://pith.science/pith/B3XVLFOHVLBZD3UL3BL6FES6IK/action/storage_attestation","attest_author":"https://pith.science/pith/B3XVLFOHVLBZD3UL3BL6FES6IK/action/author_attestation","sign_citation":"https://pith.science/pith/B3XVLFOHVLBZD3UL3BL6FES6IK/action/citation_signature","submit_replication":"https://pith.science/pith/B3XVLFOHVLBZD3UL3BL6FES6IK/action/replication_record"}},"created_at":"2026-07-05T08:36:39.452503+00:00","updated_at":"2026-07-05T08:36:39.452503+00:00"}