{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2018:NHPWD5QZV5YK53QHB3VHFXO7FM","short_pith_number":"pith:NHPWD5QZ","schema_version":"1.0","canonical_sha256":"69df61f619af70aeee070eea72dddf2b085b350814d72e7d3b1edc7d1f2672f2","source":{"kind":"arxiv","id":"1809.02322","version":2},"attestation_state":"computed","paper":{"title":"Beyond Gradient Descent for Regularized Segmentation Losses","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Dmitrii Marin, Ismail Ben Ayed, Meng Tang, Yuri Boykov","submitted_at":"2018-09-07T06:41:17Z","abstract_excerpt":"The simplicity of gradient descent (GD) made it the default method for training ever-deeper and complex neural networks. Both loss functions and architectures are often explicitly tuned to be amenable to this basic local optimization. In the context of weakly-supervised CNN segmentation, we demonstrate a well-motivated loss function where an alternative optimizer (ADM) achieves the state-of-the-art while GD performs poorly. Interestingly, GD obtains its best result for a \"smoother\" tuning of the loss function. The results are consistent across different network architectures. Our loss is motiv"},"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":"1809.02322","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-09-07T06:41:17Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"9f6b2acd2496c160965ac6b1556b21345b036909137082e3f74ba98cbb4a208d","abstract_canon_sha256":"dc356f7ba2ab1d4b060a7194746f11b39cb8423d1c4f154ea5cb14129fbae33e"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:47:36.099058Z","signature_b64":"smzVLZfTmp7Tyvd32wFozed6rhUpYLzADQTsjF06GBkYb+MY7yLP0R9jBbmdRjsN/z3CsG0eo3h0KIGn7RcwBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"69df61f619af70aeee070eea72dddf2b085b350814d72e7d3b1edc7d1f2672f2","last_reissued_at":"2026-05-17T23:47:36.098652Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:47:36.098652Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Beyond Gradient Descent for Regularized Segmentation Losses","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Dmitrii Marin, Ismail Ben Ayed, Meng Tang, Yuri Boykov","submitted_at":"2018-09-07T06:41:17Z","abstract_excerpt":"The simplicity of gradient descent (GD) made it the default method for training ever-deeper and complex neural networks. Both loss functions and architectures are often explicitly tuned to be amenable to this basic local optimization. In the context of weakly-supervised CNN segmentation, we demonstrate a well-motivated loss function where an alternative optimizer (ADM) achieves the state-of-the-art while GD performs poorly. Interestingly, GD obtains its best result for a \"smoother\" tuning of the loss function. The results are consistent across different network architectures. Our loss is motiv"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1809.02322","kind":"arxiv","version":2},"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":"1809.02322","created_at":"2026-05-17T23:47:36.098725+00:00"},{"alias_kind":"arxiv_version","alias_value":"1809.02322v2","created_at":"2026-05-17T23:47:36.098725+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1809.02322","created_at":"2026-05-17T23:47:36.098725+00:00"},{"alias_kind":"pith_short_12","alias_value":"NHPWD5QZV5YK","created_at":"2026-05-18T12:32:40.477152+00:00"},{"alias_kind":"pith_short_16","alias_value":"NHPWD5QZV5YK53QH","created_at":"2026-05-18T12:32:40.477152+00:00"},{"alias_kind":"pith_short_8","alias_value":"NHPWD5QZ","created_at":"2026-05-18T12:32:40.477152+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/NHPWD5QZV5YK53QHB3VHFXO7FM","json":"https://pith.science/pith/NHPWD5QZV5YK53QHB3VHFXO7FM.json","graph_json":"https://pith.science/api/pith-number/NHPWD5QZV5YK53QHB3VHFXO7FM/graph.json","events_json":"https://pith.science/api/pith-number/NHPWD5QZV5YK53QHB3VHFXO7FM/events.json","paper":"https://pith.science/paper/NHPWD5QZ"},"agent_actions":{"view_html":"https://pith.science/pith/NHPWD5QZV5YK53QHB3VHFXO7FM","download_json":"https://pith.science/pith/NHPWD5QZV5YK53QHB3VHFXO7FM.json","view_paper":"https://pith.science/paper/NHPWD5QZ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1809.02322&json=true","fetch_graph":"https://pith.science/api/pith-number/NHPWD5QZV5YK53QHB3VHFXO7FM/graph.json","fetch_events":"https://pith.science/api/pith-number/NHPWD5QZV5YK53QHB3VHFXO7FM/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/NHPWD5QZV5YK53QHB3VHFXO7FM/action/timestamp_anchor","attest_storage":"https://pith.science/pith/NHPWD5QZV5YK53QHB3VHFXO7FM/action/storage_attestation","attest_author":"https://pith.science/pith/NHPWD5QZV5YK53QHB3VHFXO7FM/action/author_attestation","sign_citation":"https://pith.science/pith/NHPWD5QZV5YK53QHB3VHFXO7FM/action/citation_signature","submit_replication":"https://pith.science/pith/NHPWD5QZV5YK53QHB3VHFXO7FM/action/replication_record"}},"created_at":"2026-05-17T23:47:36.098725+00:00","updated_at":"2026-05-17T23:47:36.098725+00:00"}