{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2020:NQNC2HU6GNG5JRBSYYUV7PB6JM","short_pith_number":"pith:NQNC2HU6","schema_version":"1.0","canonical_sha256":"6c1a2d1e9e334dd4c432c6295fbc3e4b223e0dc04db28d6fa050c776d728bb60","source":{"kind":"arxiv","id":"2002.01113","version":1},"attestation_state":"computed","paper":{"title":"Efficient Riemannian Optimization on the Stiefel Manifold via the Cayley Transform","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Jun Li, Li Fuxin, Sinisa Todorovic","submitted_at":"2020-02-04T04:01:51Z","abstract_excerpt":"Strictly enforcing orthonormality constraints on parameter matrices has been shown advantageous in deep learning. This amounts to Riemannian optimization on the Stiefel manifold, which, however, is computationally expensive. To address this challenge, we present two main contributions: (1) A new efficient retraction map based on an iterative Cayley transform for optimization updates, and (2) An implicit vector transport mechanism based on the combination of a projection of the momentum and the Cayley transform on the Stiefel manifold. We specify two new optimization algorithms: Cayley SGD with"},"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":"2002.01113","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2020-02-04T04:01:51Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"926467da979ac46fab5f1bfd14bc464a99924522f95644ef2d58f2b108c986ef","abstract_canon_sha256":"0a82ad65ececea8b068cd98e3def9955767ca6c3983bb58a499b630700dc9c17"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T00:38:14.822886Z","signature_b64":"3mIk6YZww2cwiO3wdSW5DtNKDpd7YznyfO3+zwuXdwr0q2TjlPPhvZf4R655H4vvTyakOfeBQLp5TnjKQfFOAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"6c1a2d1e9e334dd4c432c6295fbc3e4b223e0dc04db28d6fa050c776d728bb60","last_reissued_at":"2026-07-05T00:38:14.822417Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T00:38:14.822417Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Efficient Riemannian Optimization on the Stiefel Manifold via the Cayley Transform","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Jun Li, Li Fuxin, Sinisa Todorovic","submitted_at":"2020-02-04T04:01:51Z","abstract_excerpt":"Strictly enforcing orthonormality constraints on parameter matrices has been shown advantageous in deep learning. This amounts to Riemannian optimization on the Stiefel manifold, which, however, is computationally expensive. To address this challenge, we present two main contributions: (1) A new efficient retraction map based on an iterative Cayley transform for optimization updates, and (2) An implicit vector transport mechanism based on the combination of a projection of the momentum and the Cayley transform on the Stiefel manifold. We specify two new optimization algorithms: Cayley SGD with"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2002.01113","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/2002.01113/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":"2002.01113","created_at":"2026-07-05T00:38:14.822476+00:00"},{"alias_kind":"arxiv_version","alias_value":"2002.01113v1","created_at":"2026-07-05T00:38:14.822476+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2002.01113","created_at":"2026-07-05T00:38:14.822476+00:00"},{"alias_kind":"pith_short_12","alias_value":"NQNC2HU6GNG5","created_at":"2026-07-05T00:38:14.822476+00:00"},{"alias_kind":"pith_short_16","alias_value":"NQNC2HU6GNG5JRBS","created_at":"2026-07-05T00:38:14.822476+00:00"},{"alias_kind":"pith_short_8","alias_value":"NQNC2HU6","created_at":"2026-07-05T00:38:14.822476+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":5,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.01556","citing_title":"TwinQuant: Learnable Subspace Decomposition for 4-Bit LLM Quantization","ref_index":97,"is_internal_anchor":false},{"citing_arxiv_id":"2405.16406","citing_title":"SpinQuant: LLM quantization with learned rotations","ref_index":10,"is_internal_anchor":false},{"citing_arxiv_id":"2605.12492","citing_title":"Pion: A Spectrum-Preserving Optimizer via Orthogonal Equivalence Transformation","ref_index":41,"is_internal_anchor":false},{"citing_arxiv_id":"2604.11080","citing_title":"ReSpinQuant: Efficient Layer-Wise LLM Quantization via Subspace Residual Rotation Approximation","ref_index":4,"is_internal_anchor":false},{"citing_arxiv_id":"2502.04425","citing_title":"Tensor-Programmable Quantum Circuits for Solving Differential Equations","ref_index":98,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/NQNC2HU6GNG5JRBSYYUV7PB6JM","json":"https://pith.science/pith/NQNC2HU6GNG5JRBSYYUV7PB6JM.json","graph_json":"https://pith.science/api/pith-number/NQNC2HU6GNG5JRBSYYUV7PB6JM/graph.json","events_json":"https://pith.science/api/pith-number/NQNC2HU6GNG5JRBSYYUV7PB6JM/events.json","paper":"https://pith.science/paper/NQNC2HU6"},"agent_actions":{"view_html":"https://pith.science/pith/NQNC2HU6GNG5JRBSYYUV7PB6JM","download_json":"https://pith.science/pith/NQNC2HU6GNG5JRBSYYUV7PB6JM.json","view_paper":"https://pith.science/paper/NQNC2HU6","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2002.01113&json=true","fetch_graph":"https://pith.science/api/pith-number/NQNC2HU6GNG5JRBSYYUV7PB6JM/graph.json","fetch_events":"https://pith.science/api/pith-number/NQNC2HU6GNG5JRBSYYUV7PB6JM/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/NQNC2HU6GNG5JRBSYYUV7PB6JM/action/timestamp_anchor","attest_storage":"https://pith.science/pith/NQNC2HU6GNG5JRBSYYUV7PB6JM/action/storage_attestation","attest_author":"https://pith.science/pith/NQNC2HU6GNG5JRBSYYUV7PB6JM/action/author_attestation","sign_citation":"https://pith.science/pith/NQNC2HU6GNG5JRBSYYUV7PB6JM/action/citation_signature","submit_replication":"https://pith.science/pith/NQNC2HU6GNG5JRBSYYUV7PB6JM/action/replication_record"}},"created_at":"2026-07-05T00:38:14.822476+00:00","updated_at":"2026-07-05T00:38:14.822476+00:00"}