{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:WO6K6PLUVNMOLJBKNTTXN26RPI","short_pith_number":"pith:WO6K6PLU","schema_version":"1.0","canonical_sha256":"b3bcaf3d74ab58e5a42a6ce776ebd17a320206d6ea42f3476c7c502c09e46a0e","source":{"kind":"arxiv","id":"2311.07065","version":3},"attestation_state":"computed","paper":{"title":"On non-approximability of zero loss global ${\\mathcal L}^2$ minimizers by gradient descent in Deep Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","math-ph","math.MP","math.OC","stat.ML"],"primary_cat":"cs.LG","authors_text":"Patricia Mu\\~noz Ewald, Thomas Chen","submitted_at":"2023-11-13T04:11:25Z","abstract_excerpt":"We analyze geometric aspects of the gradient descent algorithm in Deep Learning (DL), and give a detailed discussion of the circumstance that in underparametrized DL networks, zero loss minimization can generically not be attained. As a consequence, we conclude that the distribution of training inputs must necessarily be non-generic in order to produce zero loss minimizers, both for the method constructed in [Chen-Munoz Ewald 2023, 2024], or for gradient descent [Chen 2025] (which assume clustering of training data)."},"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":"2311.07065","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2023-11-13T04:11:25Z","cross_cats_sorted":["cs.AI","math-ph","math.MP","math.OC","stat.ML"],"title_canon_sha256":"efb809716a2cf73addeef28f4bb323fa693dcc2daa1121d6878de9c7927cb5cf","abstract_canon_sha256":"92e4b40bdff1cbb037be38b9ce50c6252f26d448ee428d286b9bd75c63d9c6ef"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-09T01:05:02.973890Z","signature_b64":"ZsKgxFj8SlAs5qaSnW6Im0uFcCma61z15L35HpW9Ti9uSNpAFVJA9Sp5qLmj23rTkFtKreFfPD2pUaRONLTKDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b3bcaf3d74ab58e5a42a6ce776ebd17a320206d6ea42f3476c7c502c09e46a0e","last_reissued_at":"2026-06-09T01:05:02.973486Z","signature_status":"signed_v1","first_computed_at":"2026-06-09T01:05:02.973486Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"On non-approximability of zero loss global ${\\mathcal L}^2$ minimizers by gradient descent in Deep Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","math-ph","math.MP","math.OC","stat.ML"],"primary_cat":"cs.LG","authors_text":"Patricia Mu\\~noz Ewald, Thomas Chen","submitted_at":"2023-11-13T04:11:25Z","abstract_excerpt":"We analyze geometric aspects of the gradient descent algorithm in Deep Learning (DL), and give a detailed discussion of the circumstance that in underparametrized DL networks, zero loss minimization can generically not be attained. As a consequence, we conclude that the distribution of training inputs must necessarily be non-generic in order to produce zero loss minimizers, both for the method constructed in [Chen-Munoz Ewald 2023, 2024], or for gradient descent [Chen 2025] (which assume clustering of training data)."},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2311.07065","kind":"arxiv","version":3},"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/2311.07065/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":"2311.07065","created_at":"2026-06-09T01:05:02.973548+00:00"},{"alias_kind":"arxiv_version","alias_value":"2311.07065v3","created_at":"2026-06-09T01:05:02.973548+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2311.07065","created_at":"2026-06-09T01:05:02.973548+00:00"},{"alias_kind":"pith_short_12","alias_value":"WO6K6PLUVNMO","created_at":"2026-06-09T01:05:02.973548+00:00"},{"alias_kind":"pith_short_16","alias_value":"WO6K6PLUVNMOLJBK","created_at":"2026-06-09T01:05:02.973548+00:00"},{"alias_kind":"pith_short_8","alias_value":"WO6K6PLU","created_at":"2026-06-09T01:05:02.973548+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/WO6K6PLUVNMOLJBKNTTXN26RPI","json":"https://pith.science/pith/WO6K6PLUVNMOLJBKNTTXN26RPI.json","graph_json":"https://pith.science/api/pith-number/WO6K6PLUVNMOLJBKNTTXN26RPI/graph.json","events_json":"https://pith.science/api/pith-number/WO6K6PLUVNMOLJBKNTTXN26RPI/events.json","paper":"https://pith.science/paper/WO6K6PLU"},"agent_actions":{"view_html":"https://pith.science/pith/WO6K6PLUVNMOLJBKNTTXN26RPI","download_json":"https://pith.science/pith/WO6K6PLUVNMOLJBKNTTXN26RPI.json","view_paper":"https://pith.science/paper/WO6K6PLU","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2311.07065&json=true","fetch_graph":"https://pith.science/api/pith-number/WO6K6PLUVNMOLJBKNTTXN26RPI/graph.json","fetch_events":"https://pith.science/api/pith-number/WO6K6PLUVNMOLJBKNTTXN26RPI/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/WO6K6PLUVNMOLJBKNTTXN26RPI/action/timestamp_anchor","attest_storage":"https://pith.science/pith/WO6K6PLUVNMOLJBKNTTXN26RPI/action/storage_attestation","attest_author":"https://pith.science/pith/WO6K6PLUVNMOLJBKNTTXN26RPI/action/author_attestation","sign_citation":"https://pith.science/pith/WO6K6PLUVNMOLJBKNTTXN26RPI/action/citation_signature","submit_replication":"https://pith.science/pith/WO6K6PLUVNMOLJBKNTTXN26RPI/action/replication_record"}},"created_at":"2026-06-09T01:05:02.973548+00:00","updated_at":"2026-06-09T01:05:02.973548+00:00"}