{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2017:WKWP7U523KJTNP4Y63VE77S5PX","short_pith_number":"pith:WKWP7U52","schema_version":"1.0","canonical_sha256":"b2acffd3bada9336bf98f6ea4ffe5d7de8ca5ead80ba5acfa6588fedb19ecab4","source":{"kind":"arxiv","id":"1712.04473","version":3},"attestation_state":"computed","paper":{"title":"Enhancing approximation abilities of neural networks by training derivatives","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.NE","authors_text":"V.I. Avrutskiy","submitted_at":"2017-12-12T19:12:16Z","abstract_excerpt":"A method to increase the precision of feedforward networks is proposed. It requires a prior knowledge of a target function derivatives of several orders and uses this information in gradient based training. Forward pass calculates not only the values of the output layer of a network but also their derivatives. The deviations of those derivatives from the target ones are used in an extended cost function and then backward pass calculates the gradient of the extended cost with respect to weights, which can then be used by any weights update algorithm. Despite a substantial increase in arithmetic"},"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":"1712.04473","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.NE","submitted_at":"2017-12-12T19:12:16Z","cross_cats_sorted":[],"title_canon_sha256":"7d2b405e9907d38825baef4846b996d15d51aed0019e773d44b047adcd712e97","abstract_canon_sha256":"180278b505086ddbb24b5a272227dfbb92157ce5bd414be5f779a0571efdf598"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T00:53:04.330879Z","signature_b64":"sfGGFrUYMk4RFS8IlIhnJrW19N/8Z/6dtN6IqV+Pr7rkc83c0J4H969KCYuj37Y9cwog+dUpAaWsfqpHtLCRDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b2acffd3bada9336bf98f6ea4ffe5d7de8ca5ead80ba5acfa6588fedb19ecab4","last_reissued_at":"2026-07-05T00:53:04.330468Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T00:53:04.330468Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Enhancing approximation abilities of neural networks by training derivatives","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.NE","authors_text":"V.I. Avrutskiy","submitted_at":"2017-12-12T19:12:16Z","abstract_excerpt":"A method to increase the precision of feedforward networks is proposed. It requires a prior knowledge of a target function derivatives of several orders and uses this information in gradient based training. Forward pass calculates not only the values of the output layer of a network but also their derivatives. The deviations of those derivatives from the target ones are used in an extended cost function and then backward pass calculates the gradient of the extended cost with respect to weights, which can then be used by any weights update algorithm. Despite a substantial increase in arithmetic"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1712.04473","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/1712.04473/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":"1712.04473","created_at":"2026-07-05T00:53:04.330533+00:00"},{"alias_kind":"arxiv_version","alias_value":"1712.04473v3","created_at":"2026-07-05T00:53:04.330533+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1712.04473","created_at":"2026-07-05T00:53:04.330533+00:00"},{"alias_kind":"pith_short_12","alias_value":"WKWP7U523KJT","created_at":"2026-07-05T00:53:04.330533+00:00"},{"alias_kind":"pith_short_16","alias_value":"WKWP7U523KJTNP4Y","created_at":"2026-07-05T00:53:04.330533+00:00"},{"alias_kind":"pith_short_8","alias_value":"WKWP7U52","created_at":"2026-07-05T00:53:04.330533+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"1908.10407","citing_title":"An Energy Approach to the Solution of Partial Differential Equations in Computational Mechanics via Machine Learning: Concepts, Implementation and Applications","ref_index":24,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/WKWP7U523KJTNP4Y63VE77S5PX","json":"https://pith.science/pith/WKWP7U523KJTNP4Y63VE77S5PX.json","graph_json":"https://pith.science/api/pith-number/WKWP7U523KJTNP4Y63VE77S5PX/graph.json","events_json":"https://pith.science/api/pith-number/WKWP7U523KJTNP4Y63VE77S5PX/events.json","paper":"https://pith.science/paper/WKWP7U52"},"agent_actions":{"view_html":"https://pith.science/pith/WKWP7U523KJTNP4Y63VE77S5PX","download_json":"https://pith.science/pith/WKWP7U523KJTNP4Y63VE77S5PX.json","view_paper":"https://pith.science/paper/WKWP7U52","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1712.04473&json=true","fetch_graph":"https://pith.science/api/pith-number/WKWP7U523KJTNP4Y63VE77S5PX/graph.json","fetch_events":"https://pith.science/api/pith-number/WKWP7U523KJTNP4Y63VE77S5PX/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/WKWP7U523KJTNP4Y63VE77S5PX/action/timestamp_anchor","attest_storage":"https://pith.science/pith/WKWP7U523KJTNP4Y63VE77S5PX/action/storage_attestation","attest_author":"https://pith.science/pith/WKWP7U523KJTNP4Y63VE77S5PX/action/author_attestation","sign_citation":"https://pith.science/pith/WKWP7U523KJTNP4Y63VE77S5PX/action/citation_signature","submit_replication":"https://pith.science/pith/WKWP7U523KJTNP4Y63VE77S5PX/action/replication_record"}},"created_at":"2026-07-05T00:53:04.330533+00:00","updated_at":"2026-07-05T00:53:04.330533+00:00"}