{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:QWU6VBVMIUPPQPPOCVPIAJWOJT","short_pith_number":"pith:QWU6VBVM","schema_version":"1.0","canonical_sha256":"85a9ea86ac451ef83dee155e8026ce4cffe2a88ec495430ae086474efd4d1844","source":{"kind":"arxiv","id":"2305.18078","version":1},"attestation_state":"computed","paper":{"title":"The mechanism underlying successful deep learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Ido Kanter, Ofek Tevet, Ronit D. Gross, Roni Vardi, Shiri Hodassman, Yarden Tzach, Yuval Meir","submitted_at":"2023-05-29T13:28:43Z","abstract_excerpt":"Deep architectures consist of tens or hundreds of convolutional layers (CLs) that terminate with a few fully connected (FC) layers and an output layer representing the possible labels of a complex classification task. According to the existing deep learning (DL) rationale, the first CL reveals localized features from the raw data, whereas the subsequent layers progressively extract higher-level features required for refined classification. This article presents an efficient three-phase procedure for quantifying the mechanism underlying successful DL. First, a deep architecture is trained to ma"},"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":"2305.18078","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2023-05-29T13:28:43Z","cross_cats_sorted":[],"title_canon_sha256":"26edc378591a351b9cec0e3ec7cacf4a8918199d4c75c0fa921e5a94216fb904","abstract_canon_sha256":"bd60674d2e4f45691b38b975835ebec9e497cb7ea711b3a8caaa0c59924b89ce"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:14:52.285189Z","signature_b64":"dx5yvL5prep7Vbi3T6bokoH08B51FXShe5Wsvxbt19XnTeBo1d8VR6HSxC8ZdVI3Zu6MzoRjdoA+nuCJWpgpCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"85a9ea86ac451ef83dee155e8026ce4cffe2a88ec495430ae086474efd4d1844","last_reissued_at":"2026-07-05T06:14:52.284764Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:14:52.284764Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"The mechanism underlying successful deep learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Ido Kanter, Ofek Tevet, Ronit D. Gross, Roni Vardi, Shiri Hodassman, Yarden Tzach, Yuval Meir","submitted_at":"2023-05-29T13:28:43Z","abstract_excerpt":"Deep architectures consist of tens or hundreds of convolutional layers (CLs) that terminate with a few fully connected (FC) layers and an output layer representing the possible labels of a complex classification task. According to the existing deep learning (DL) rationale, the first CL reveals localized features from the raw data, whereas the subsequent layers progressively extract higher-level features required for refined classification. This article presents an efficient three-phase procedure for quantifying the mechanism underlying successful DL. First, a deep architecture is trained to ma"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2305.18078","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/2305.18078/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":"2305.18078","created_at":"2026-07-05T06:14:52.284820+00:00"},{"alias_kind":"arxiv_version","alias_value":"2305.18078v1","created_at":"2026-07-05T06:14:52.284820+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2305.18078","created_at":"2026-07-05T06:14:52.284820+00:00"},{"alias_kind":"pith_short_12","alias_value":"QWU6VBVMIUPP","created_at":"2026-07-05T06:14:52.284820+00:00"},{"alias_kind":"pith_short_16","alias_value":"QWU6VBVMIUPPQPPO","created_at":"2026-07-05T06:14:52.284820+00:00"},{"alias_kind":"pith_short_8","alias_value":"QWU6VBVM","created_at":"2026-07-05T06:14:52.284820+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2509.03407","citing_title":"Learning Mechanism Underlying NLP Pre-Training and Fine-Tuning","ref_index":41,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/QWU6VBVMIUPPQPPOCVPIAJWOJT","json":"https://pith.science/pith/QWU6VBVMIUPPQPPOCVPIAJWOJT.json","graph_json":"https://pith.science/api/pith-number/QWU6VBVMIUPPQPPOCVPIAJWOJT/graph.json","events_json":"https://pith.science/api/pith-number/QWU6VBVMIUPPQPPOCVPIAJWOJT/events.json","paper":"https://pith.science/paper/QWU6VBVM"},"agent_actions":{"view_html":"https://pith.science/pith/QWU6VBVMIUPPQPPOCVPIAJWOJT","download_json":"https://pith.science/pith/QWU6VBVMIUPPQPPOCVPIAJWOJT.json","view_paper":"https://pith.science/paper/QWU6VBVM","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2305.18078&json=true","fetch_graph":"https://pith.science/api/pith-number/QWU6VBVMIUPPQPPOCVPIAJWOJT/graph.json","fetch_events":"https://pith.science/api/pith-number/QWU6VBVMIUPPQPPOCVPIAJWOJT/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/QWU6VBVMIUPPQPPOCVPIAJWOJT/action/timestamp_anchor","attest_storage":"https://pith.science/pith/QWU6VBVMIUPPQPPOCVPIAJWOJT/action/storage_attestation","attest_author":"https://pith.science/pith/QWU6VBVMIUPPQPPOCVPIAJWOJT/action/author_attestation","sign_citation":"https://pith.science/pith/QWU6VBVMIUPPQPPOCVPIAJWOJT/action/citation_signature","submit_replication":"https://pith.science/pith/QWU6VBVMIUPPQPPOCVPIAJWOJT/action/replication_record"}},"created_at":"2026-07-05T06:14:52.284820+00:00","updated_at":"2026-07-05T06:14:52.284820+00:00"}