{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2018:RG22BSLNAJOCJ6LLEM7T2NRWBP","short_pith_number":"pith:RG22BSLN","schema_version":"1.0","canonical_sha256":"89b5a0c96d025c24f96b233f3d36360bc341d194cf805f6ffed298579f3e1c62","source":{"kind":"arxiv","id":"1802.07042","version":3},"attestation_state":"computed","paper":{"title":"Do deep nets really need weight decay and dropout?","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Alex Hern\\'andez-Garc\\'ia, Peter K\\\"onig","submitted_at":"2018-02-20T10:12:23Z","abstract_excerpt":"The impressive success of modern deep neural networks on computer vision tasks has been achieved through models of very large capacity compared to the number of available training examples. This overparameterization is often said to be controlled with the help of different regularization techniques, mainly weight decay and dropout. However, since these techniques reduce the effective capacity of the model, typically even deeper and wider architectures are required to compensate for the reduced capacity. Therefore, there seems to be a waste of capacity in this practice. In this paper we build u"},"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":"1802.07042","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2018-02-20T10:12:23Z","cross_cats_sorted":[],"title_canon_sha256":"1c1922a724659bfa4aea27f3426623b803635a9b30d8bd4561dd22b1bdf35f52","abstract_canon_sha256":"8916e3bb38ec71ce5bd27347ed73a097e5df373ce6f01d289a84071ad7fb08d9"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T00:10:54.557942Z","signature_b64":"e6efsx7Ln4BVgnf7689YcTB9Wqe1A1saMdNjGhY8cJCUjWZraLdU8Tf/niKxwbOX7uO6DkABo1h+wKymrAQ5Dg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"89b5a0c96d025c24f96b233f3d36360bc341d194cf805f6ffed298579f3e1c62","last_reissued_at":"2026-05-18T00:10:54.557125Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T00:10:54.557125Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Do deep nets really need weight decay and dropout?","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Alex Hern\\'andez-Garc\\'ia, Peter K\\\"onig","submitted_at":"2018-02-20T10:12:23Z","abstract_excerpt":"The impressive success of modern deep neural networks on computer vision tasks has been achieved through models of very large capacity compared to the number of available training examples. This overparameterization is often said to be controlled with the help of different regularization techniques, mainly weight decay and dropout. However, since these techniques reduce the effective capacity of the model, typically even deeper and wider architectures are required to compensate for the reduced capacity. Therefore, there seems to be a waste of capacity in this practice. In this paper we build u"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1802.07042","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":""},"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":"1802.07042","created_at":"2026-05-18T00:10:54.557262+00:00"},{"alias_kind":"arxiv_version","alias_value":"1802.07042v3","created_at":"2026-05-18T00:10:54.557262+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1802.07042","created_at":"2026-05-18T00:10:54.557262+00:00"},{"alias_kind":"pith_short_12","alias_value":"RG22BSLNAJOC","created_at":"2026-05-18T12:32:50.500415+00:00"},{"alias_kind":"pith_short_16","alias_value":"RG22BSLNAJOCJ6LL","created_at":"2026-05-18T12:32:50.500415+00:00"},{"alias_kind":"pith_short_8","alias_value":"RG22BSLN","created_at":"2026-05-18T12:32:50.500415+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"1906.11052","citing_title":"Further advantages of data augmentation on convolutional neural networks","ref_index":16,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/RG22BSLNAJOCJ6LLEM7T2NRWBP","json":"https://pith.science/pith/RG22BSLNAJOCJ6LLEM7T2NRWBP.json","graph_json":"https://pith.science/api/pith-number/RG22BSLNAJOCJ6LLEM7T2NRWBP/graph.json","events_json":"https://pith.science/api/pith-number/RG22BSLNAJOCJ6LLEM7T2NRWBP/events.json","paper":"https://pith.science/paper/RG22BSLN"},"agent_actions":{"view_html":"https://pith.science/pith/RG22BSLNAJOCJ6LLEM7T2NRWBP","download_json":"https://pith.science/pith/RG22BSLNAJOCJ6LLEM7T2NRWBP.json","view_paper":"https://pith.science/paper/RG22BSLN","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1802.07042&json=true","fetch_graph":"https://pith.science/api/pith-number/RG22BSLNAJOCJ6LLEM7T2NRWBP/graph.json","fetch_events":"https://pith.science/api/pith-number/RG22BSLNAJOCJ6LLEM7T2NRWBP/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/RG22BSLNAJOCJ6LLEM7T2NRWBP/action/timestamp_anchor","attest_storage":"https://pith.science/pith/RG22BSLNAJOCJ6LLEM7T2NRWBP/action/storage_attestation","attest_author":"https://pith.science/pith/RG22BSLNAJOCJ6LLEM7T2NRWBP/action/author_attestation","sign_citation":"https://pith.science/pith/RG22BSLNAJOCJ6LLEM7T2NRWBP/action/citation_signature","submit_replication":"https://pith.science/pith/RG22BSLNAJOCJ6LLEM7T2NRWBP/action/replication_record"}},"created_at":"2026-05-18T00:10:54.557262+00:00","updated_at":"2026-05-18T00:10:54.557262+00:00"}