{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:7TTBDDKATLHNFJPTRO6OHMPV52","short_pith_number":"pith:7TTBDDKA","schema_version":"1.0","canonical_sha256":"fce6118d409aced2a5f38bbce3b1f5ee8c28b00154a9c914af5a5a1c8575af1c","source":{"kind":"arxiv","id":"2404.14855","version":1},"attestation_state":"computed","paper":{"title":"The Geometry of the Set of Equivalent Linear Neural Networks","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.CG"],"primary_cat":"cs.LG","authors_text":"Jonathan Richard Shewchuk, Sagnik Bhattacharya","submitted_at":"2024-04-23T09:20:55Z","abstract_excerpt":"We characterize the geometry and topology of the set of all weight vectors for which a linear neural network computes the same linear transformation $W$. This set of weight vectors is called the fiber of $W$ (under the matrix multiplication map), and it is embedded in the Euclidean weight space of all possible weight vectors. The fiber is an algebraic variety that is not necessarily a manifold. We describe a natural way to stratify the fiber--that is, to partition the algebraic variety into a finite set of manifolds of varying dimensions called strata. We call this set of strata the rank strat"},"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":"2404.14855","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.LG","submitted_at":"2024-04-23T09:20:55Z","cross_cats_sorted":["cs.CG"],"title_canon_sha256":"179ee9e65509ac22bd49c6fb33c0b755cac690b094db00803b30f121e211ecc4","abstract_canon_sha256":"db71ce9b6a5b12b45c08f9879b9e795593cc8601bad3da104e54a49da59df647"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:11:11.139679Z","signature_b64":"F1mVwlNcu2OxcLF8qmZ1F9lLq43wWKRor7KG2kkXAjEP7sp58Kpn4muKJ80el1s/fD0ErmIIv+RlyPXVmvVxBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"fce6118d409aced2a5f38bbce3b1f5ee8c28b00154a9c914af5a5a1c8575af1c","last_reissued_at":"2026-07-05T08:11:11.139187Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:11:11.139187Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"The Geometry of the Set of Equivalent Linear Neural Networks","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.CG"],"primary_cat":"cs.LG","authors_text":"Jonathan Richard Shewchuk, Sagnik Bhattacharya","submitted_at":"2024-04-23T09:20:55Z","abstract_excerpt":"We characterize the geometry and topology of the set of all weight vectors for which a linear neural network computes the same linear transformation $W$. This set of weight vectors is called the fiber of $W$ (under the matrix multiplication map), and it is embedded in the Euclidean weight space of all possible weight vectors. The fiber is an algebraic variety that is not necessarily a manifold. We describe a natural way to stratify the fiber--that is, to partition the algebraic variety into a finite set of manifolds of varying dimensions called strata. We call this set of strata the rank strat"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2404.14855","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/2404.14855/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":"2404.14855","created_at":"2026-07-05T08:11:11.139259+00:00"},{"alias_kind":"arxiv_version","alias_value":"2404.14855v1","created_at":"2026-07-05T08:11:11.139259+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2404.14855","created_at":"2026-07-05T08:11:11.139259+00:00"},{"alias_kind":"pith_short_12","alias_value":"7TTBDDKATLHN","created_at":"2026-07-05T08:11:11.139259+00:00"},{"alias_kind":"pith_short_16","alias_value":"7TTBDDKATLHNFJPT","created_at":"2026-07-05T08:11:11.139259+00:00"},{"alias_kind":"pith_short_8","alias_value":"7TTBDDKA","created_at":"2026-07-05T08:11:11.139259+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2411.19920","citing_title":"Geometry of fibers of the multiplication map of deep linear neural networks","ref_index":39,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/7TTBDDKATLHNFJPTRO6OHMPV52","json":"https://pith.science/pith/7TTBDDKATLHNFJPTRO6OHMPV52.json","graph_json":"https://pith.science/api/pith-number/7TTBDDKATLHNFJPTRO6OHMPV52/graph.json","events_json":"https://pith.science/api/pith-number/7TTBDDKATLHNFJPTRO6OHMPV52/events.json","paper":"https://pith.science/paper/7TTBDDKA"},"agent_actions":{"view_html":"https://pith.science/pith/7TTBDDKATLHNFJPTRO6OHMPV52","download_json":"https://pith.science/pith/7TTBDDKATLHNFJPTRO6OHMPV52.json","view_paper":"https://pith.science/paper/7TTBDDKA","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2404.14855&json=true","fetch_graph":"https://pith.science/api/pith-number/7TTBDDKATLHNFJPTRO6OHMPV52/graph.json","fetch_events":"https://pith.science/api/pith-number/7TTBDDKATLHNFJPTRO6OHMPV52/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/7TTBDDKATLHNFJPTRO6OHMPV52/action/timestamp_anchor","attest_storage":"https://pith.science/pith/7TTBDDKATLHNFJPTRO6OHMPV52/action/storage_attestation","attest_author":"https://pith.science/pith/7TTBDDKATLHNFJPTRO6OHMPV52/action/author_attestation","sign_citation":"https://pith.science/pith/7TTBDDKATLHNFJPTRO6OHMPV52/action/citation_signature","submit_replication":"https://pith.science/pith/7TTBDDKATLHNFJPTRO6OHMPV52/action/replication_record"}},"created_at":"2026-07-05T08:11:11.139259+00:00","updated_at":"2026-07-05T08:11:11.139259+00:00"}