{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2019:EXIQPPVPG2Y3AC44KZMSXUPP4K","short_pith_number":"pith:EXIQPPVP","schema_version":"1.0","canonical_sha256":"25d107beaf36b1b00b9c56592bd1efe2a9a3911467c234af57cc02cef6fbc327","source":{"kind":"arxiv","id":"1903.01182","version":2},"attestation_state":"computed","paper":{"title":"Complement Objective Training","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CV","stat.ML"],"primary_cat":"cs.LG","authors_text":"Chun-Hao Liu, Da-Cheng Juan, Hao-Yun Chen, Jia-Yu Pan, Pei-Hsin Wang, Shih-Chieh Chang, Wei Wei, Yu-Ting Chen","submitted_at":"2019-03-04T11:35:27Z","abstract_excerpt":"Learning with a primary objective, such as softmax cross entropy for classification and sequence generation, has been the norm for training deep neural networks for years. Although being a widely-adopted approach, using cross entropy as the primary objective exploits mostly the information from the ground-truth class for maximizing data likelihood, and largely ignores information from the complement (incorrect) classes. We argue that, in addition to the primary objective, training also using a complement objective that leverages information from the complement classes can be effective in impro"},"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":"1903.01182","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-03-04T11:35:27Z","cross_cats_sorted":["cs.CV","stat.ML"],"title_canon_sha256":"df7e1dacf1c2ee06e33d13579f2be87a6f23c81ea6664224be527ea2276461ab","abstract_canon_sha256":"a6025f454fe1c6c2c8abd7cc94ba37fc9ebaa72eb6215ba9826bfc555c2e86ec"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:50:40.601525Z","signature_b64":"Im+UrRIKRpV7QBJksUKLSSb0aSX234xlbtzt50fMeaDUqdCbhBxaDfY6JvlSC2YCHvV+QyD1f0AenErnd4x/CA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"25d107beaf36b1b00b9c56592bd1efe2a9a3911467c234af57cc02cef6fbc327","last_reissued_at":"2026-05-17T23:50:40.600829Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:50:40.600829Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Complement Objective Training","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CV","stat.ML"],"primary_cat":"cs.LG","authors_text":"Chun-Hao Liu, Da-Cheng Juan, Hao-Yun Chen, Jia-Yu Pan, Pei-Hsin Wang, Shih-Chieh Chang, Wei Wei, Yu-Ting Chen","submitted_at":"2019-03-04T11:35:27Z","abstract_excerpt":"Learning with a primary objective, such as softmax cross entropy for classification and sequence generation, has been the norm for training deep neural networks for years. Although being a widely-adopted approach, using cross entropy as the primary objective exploits mostly the information from the ground-truth class for maximizing data likelihood, and largely ignores information from the complement (incorrect) classes. We argue that, in addition to the primary objective, training also using a complement objective that leverages information from the complement classes can be effective in impro"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1903.01182","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":"1903.01182","created_at":"2026-05-17T23:50:40.600934+00:00"},{"alias_kind":"arxiv_version","alias_value":"1903.01182v2","created_at":"2026-05-17T23:50:40.600934+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1903.01182","created_at":"2026-05-17T23:50:40.600934+00:00"},{"alias_kind":"pith_short_12","alias_value":"EXIQPPVPG2Y3","created_at":"2026-05-18T12:33:15.570797+00:00"},{"alias_kind":"pith_short_16","alias_value":"EXIQPPVPG2Y3AC44","created_at":"2026-05-18T12:33:15.570797+00:00"},{"alias_kind":"pith_short_8","alias_value":"EXIQPPVP","created_at":"2026-05-18T12:33:15.570797+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/EXIQPPVPG2Y3AC44KZMSXUPP4K","json":"https://pith.science/pith/EXIQPPVPG2Y3AC44KZMSXUPP4K.json","graph_json":"https://pith.science/api/pith-number/EXIQPPVPG2Y3AC44KZMSXUPP4K/graph.json","events_json":"https://pith.science/api/pith-number/EXIQPPVPG2Y3AC44KZMSXUPP4K/events.json","paper":"https://pith.science/paper/EXIQPPVP"},"agent_actions":{"view_html":"https://pith.science/pith/EXIQPPVPG2Y3AC44KZMSXUPP4K","download_json":"https://pith.science/pith/EXIQPPVPG2Y3AC44KZMSXUPP4K.json","view_paper":"https://pith.science/paper/EXIQPPVP","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1903.01182&json=true","fetch_graph":"https://pith.science/api/pith-number/EXIQPPVPG2Y3AC44KZMSXUPP4K/graph.json","fetch_events":"https://pith.science/api/pith-number/EXIQPPVPG2Y3AC44KZMSXUPP4K/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/EXIQPPVPG2Y3AC44KZMSXUPP4K/action/timestamp_anchor","attest_storage":"https://pith.science/pith/EXIQPPVPG2Y3AC44KZMSXUPP4K/action/storage_attestation","attest_author":"https://pith.science/pith/EXIQPPVPG2Y3AC44KZMSXUPP4K/action/author_attestation","sign_citation":"https://pith.science/pith/EXIQPPVPG2Y3AC44KZMSXUPP4K/action/citation_signature","submit_replication":"https://pith.science/pith/EXIQPPVPG2Y3AC44KZMSXUPP4K/action/replication_record"}},"created_at":"2026-05-17T23:50:40.600934+00:00","updated_at":"2026-05-17T23:50:40.600934+00:00"}