{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:HM7AEPVPGFRNFJMOB5RQPAJIG6","short_pith_number":"pith:HM7AEPVP","schema_version":"1.0","canonical_sha256":"3b3e023eaf3162d2a58e0f6307812837b9938ad4e6c44c76154665824ca1b5f8","source":{"kind":"arxiv","id":"2608.11020","version":1},"attestation_state":"computed","paper":{"title":"Derivative Computation in PINNs: Automatic Differentiation, Finite Differences and Beyond","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.NA","math.NA","physics.comp-ph"],"primary_cat":"cs.LG","authors_text":"Maciej J. Mikulski, Tadeusz Uhl","submitted_at":"2026-08-11T15:01:29Z","abstract_excerpt":"We systematically investigate finite-difference (FD) derivative computation in Physics-Informed Neural Networks (PINNs) as an alternative to automatic differentiation (AD). On three benchmark PDEs we show that, with a properly calibrated step size, FD matches AD in accuracy on every problem while running faster across the full tested batch-size range and using substantially less GPU memory, and that a stochastic variant we propose outperforms AD on a stationary problem. We further show that for neural architectures with inter-sample dependencies (e.g. BatchNorm, self-attention) the standard Py"},"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":"2608.11020","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-08-11T15:01:29Z","cross_cats_sorted":["cs.NA","math.NA","physics.comp-ph"],"title_canon_sha256":"e44c18115bd16b6b66f4260de417de45b5376f09432de12406536b94ec664865","abstract_canon_sha256":"b0b016e3cc3a72144c25251c45abdae47b50a0f742711c68f9f119f6fdf8d8f5"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-08-12T01:24:05.096607Z","signature_b64":"B6AF1zgNR2isoMCgy4DV9U9jFNm9y8dkRfPJfJTBwHJWdRYtdxbggG38ErHMQSUFWiff493YAHZHwJomjlzQBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3b3e023eaf3162d2a58e0f6307812837b9938ad4e6c44c76154665824ca1b5f8","last_reissued_at":"2026-08-12T01:24:05.094847Z","signature_status":"signed_v1","first_computed_at":"2026-08-12T01:24:05.094847Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Derivative Computation in PINNs: Automatic Differentiation, Finite Differences and Beyond","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.NA","math.NA","physics.comp-ph"],"primary_cat":"cs.LG","authors_text":"Maciej J. Mikulski, Tadeusz Uhl","submitted_at":"2026-08-11T15:01:29Z","abstract_excerpt":"We systematically investigate finite-difference (FD) derivative computation in Physics-Informed Neural Networks (PINNs) as an alternative to automatic differentiation (AD). On three benchmark PDEs we show that, with a properly calibrated step size, FD matches AD in accuracy on every problem while running faster across the full tested batch-size range and using substantially less GPU memory, and that a stochastic variant we propose outperforms AD on a stationary problem. We further show that for neural architectures with inter-sample dependencies (e.g. BatchNorm, self-attention) the standard Py"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2608.11020","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/2608.11020/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":"2608.11020","created_at":"2026-08-12T01:24:05.098854+00:00"},{"alias_kind":"arxiv_version","alias_value":"2608.11020v1","created_at":"2026-08-12T01:24:05.098854+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2608.11020","created_at":"2026-08-12T01:24:05.098854+00:00"},{"alias_kind":"pith_short_12","alias_value":"HM7AEPVPGFRN","created_at":"2026-08-12T01:24:05.098854+00:00"},{"alias_kind":"pith_short_16","alias_value":"HM7AEPVPGFRNFJMO","created_at":"2026-08-12T01:24:05.098854+00:00"},{"alias_kind":"pith_short_8","alias_value":"HM7AEPVP","created_at":"2026-08-12T01:24:05.098854+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/HM7AEPVPGFRNFJMOB5RQPAJIG6","json":"https://pith.science/pith/HM7AEPVPGFRNFJMOB5RQPAJIG6.json","graph_json":"https://pith.science/api/pith-number/HM7AEPVPGFRNFJMOB5RQPAJIG6/graph.json","events_json":"https://pith.science/api/pith-number/HM7AEPVPGFRNFJMOB5RQPAJIG6/events.json","paper":"https://pith.science/paper/HM7AEPVP"},"agent_actions":{"view_html":"https://pith.science/pith/HM7AEPVPGFRNFJMOB5RQPAJIG6","download_json":"https://pith.science/pith/HM7AEPVPGFRNFJMOB5RQPAJIG6.json","view_paper":"https://pith.science/paper/HM7AEPVP","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2608.11020&json=true","fetch_graph":"https://pith.science/api/pith-number/HM7AEPVPGFRNFJMOB5RQPAJIG6/graph.json","fetch_events":"https://pith.science/api/pith-number/HM7AEPVPGFRNFJMOB5RQPAJIG6/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/HM7AEPVPGFRNFJMOB5RQPAJIG6/action/timestamp_anchor","attest_storage":"https://pith.science/pith/HM7AEPVPGFRNFJMOB5RQPAJIG6/action/storage_attestation","attest_author":"https://pith.science/pith/HM7AEPVPGFRNFJMOB5RQPAJIG6/action/author_attestation","sign_citation":"https://pith.science/pith/HM7AEPVPGFRNFJMOB5RQPAJIG6/action/citation_signature","submit_replication":"https://pith.science/pith/HM7AEPVPGFRNFJMOB5RQPAJIG6/action/replication_record"}},"created_at":"2026-08-12T01:24:05.098854+00:00","updated_at":"2026-08-12T01:24:05.098854+00:00"}