{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:5K3RNSZ7IWHISTQOM4W5PIJPAI","short_pith_number":"pith:5K3RNSZ7","schema_version":"1.0","canonical_sha256":"eab716cb3f458e894e0e672dd7a12f020b0765c53dd2bd529f2972736268dba8","source":{"kind":"arxiv","id":"2402.00531","version":1},"attestation_state":"computed","paper":{"title":"Preconditioning for Physics-Informed Neural Networks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.NA","math.NA"],"primary_cat":"cs.LG","authors_text":"Chang Su, Hang Su, Jiachen Yao, Jun Zhu, Songming Liu, Youjia Wu, Zhongkai Hao","submitted_at":"2024-02-01T11:58:28Z","abstract_excerpt":"Physics-informed neural networks (PINNs) have shown promise in solving various partial differential equations (PDEs). However, training pathologies have negatively affected the convergence and prediction accuracy of PINNs, which further limits their practical applications. In this paper, we propose to use condition number as a metric to diagnose and mitigate the pathologies in PINNs. Inspired by classical numerical analysis, where the condition number measures sensitivity and stability, we highlight its pivotal role in the training dynamics of PINNs. We prove theorems to reveal how condition n"},"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":"2402.00531","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-02-01T11:58:28Z","cross_cats_sorted":["cs.NA","math.NA"],"title_canon_sha256":"d499358a9eee4b1b34605a97b6c374ddbe55d7caeb8c4b2e7d78736856787b82","abstract_canon_sha256":"16c24c4b9ee6a282a4d217dcfd78bf296ad885927115cb882600b661a5160bcc"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:40:13.086142Z","signature_b64":"TW2Eal/0TCosF0BSEt+HbnEH3r1fJuj7YjspOW2ry/SPFdf2xk22bMFi6FlBl+XlKwVrsCIgGCXpvxm7G0WACg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"eab716cb3f458e894e0e672dd7a12f020b0765c53dd2bd529f2972736268dba8","last_reissued_at":"2026-07-05T07:40:13.085659Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:40:13.085659Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Preconditioning for Physics-Informed Neural Networks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.NA","math.NA"],"primary_cat":"cs.LG","authors_text":"Chang Su, Hang Su, Jiachen Yao, Jun Zhu, Songming Liu, Youjia Wu, Zhongkai Hao","submitted_at":"2024-02-01T11:58:28Z","abstract_excerpt":"Physics-informed neural networks (PINNs) have shown promise in solving various partial differential equations (PDEs). However, training pathologies have negatively affected the convergence and prediction accuracy of PINNs, which further limits their practical applications. In this paper, we propose to use condition number as a metric to diagnose and mitigate the pathologies in PINNs. Inspired by classical numerical analysis, where the condition number measures sensitivity and stability, we highlight its pivotal role in the training dynamics of PINNs. We prove theorems to reveal how condition n"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2402.00531","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/2402.00531/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":"2402.00531","created_at":"2026-07-05T07:40:13.085722+00:00"},{"alias_kind":"arxiv_version","alias_value":"2402.00531v1","created_at":"2026-07-05T07:40:13.085722+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2402.00531","created_at":"2026-07-05T07:40:13.085722+00:00"},{"alias_kind":"pith_short_12","alias_value":"5K3RNSZ7IWHI","created_at":"2026-07-05T07:40:13.085722+00:00"},{"alias_kind":"pith_short_16","alias_value":"5K3RNSZ7IWHISTQO","created_at":"2026-07-05T07:40:13.085722+00:00"},{"alias_kind":"pith_short_8","alias_value":"5K3RNSZ7","created_at":"2026-07-05T07:40:13.085722+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.30910","citing_title":"PINNs Failure Modes are Overfitting","ref_index":19,"is_internal_anchor":false},{"citing_arxiv_id":"2605.07286","citing_title":"Sparse Random-Feature Neural Networks with Krylov-Based SVD for Singularly Perturbed ODE","ref_index":28,"is_internal_anchor":false},{"citing_arxiv_id":"2605.07365","citing_title":"Solving Convolution-type Integral Equations using Preconditioned Neural Operators","ref_index":10,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/5K3RNSZ7IWHISTQOM4W5PIJPAI","json":"https://pith.science/pith/5K3RNSZ7IWHISTQOM4W5PIJPAI.json","graph_json":"https://pith.science/api/pith-number/5K3RNSZ7IWHISTQOM4W5PIJPAI/graph.json","events_json":"https://pith.science/api/pith-number/5K3RNSZ7IWHISTQOM4W5PIJPAI/events.json","paper":"https://pith.science/paper/5K3RNSZ7"},"agent_actions":{"view_html":"https://pith.science/pith/5K3RNSZ7IWHISTQOM4W5PIJPAI","download_json":"https://pith.science/pith/5K3RNSZ7IWHISTQOM4W5PIJPAI.json","view_paper":"https://pith.science/paper/5K3RNSZ7","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2402.00531&json=true","fetch_graph":"https://pith.science/api/pith-number/5K3RNSZ7IWHISTQOM4W5PIJPAI/graph.json","fetch_events":"https://pith.science/api/pith-number/5K3RNSZ7IWHISTQOM4W5PIJPAI/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/5K3RNSZ7IWHISTQOM4W5PIJPAI/action/timestamp_anchor","attest_storage":"https://pith.science/pith/5K3RNSZ7IWHISTQOM4W5PIJPAI/action/storage_attestation","attest_author":"https://pith.science/pith/5K3RNSZ7IWHISTQOM4W5PIJPAI/action/author_attestation","sign_citation":"https://pith.science/pith/5K3RNSZ7IWHISTQOM4W5PIJPAI/action/citation_signature","submit_replication":"https://pith.science/pith/5K3RNSZ7IWHISTQOM4W5PIJPAI/action/replication_record"}},"created_at":"2026-07-05T07:40:13.085722+00:00","updated_at":"2026-07-05T07:40:13.085722+00:00"}