{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2019:5UAVUI45ZDXNFJRR77M3WP56LG","short_pith_number":"pith:5UAVUI45","schema_version":"1.0","canonical_sha256":"ed015a239dc8eed2a631ffd9bb3fbe59b86c840ba623558a426e794b24561757","source":{"kind":"arxiv","id":"1910.06943","version":2},"attestation_state":"computed","paper":{"title":"The Local Elasticity of Neural Networks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CV","stat.ML"],"primary_cat":"cs.LG","authors_text":"Hangfeng He, Weijie J. Su","submitted_at":"2019-10-15T17:35:30Z","abstract_excerpt":"This paper presents a phenomenon in neural networks that we refer to as \\textit{local elasticity}. Roughly speaking, a classifier is said to be locally elastic if its prediction at a feature vector $\\bx'$ is \\textit{not} significantly perturbed, after the classifier is updated via stochastic gradient descent at a (labeled) feature vector $\\bx$ that is \\textit{dissimilar} to $\\bx'$ in a certain sense. This phenomenon is shown to persist for neural networks with nonlinear activation functions through extensive simulations on real-life and synthetic datasets, whereas this is not observed in linea"},"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":"1910.06943","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-10-15T17:35:30Z","cross_cats_sorted":["cs.CV","stat.ML"],"title_canon_sha256":"3eb286475884e511e21d4b7d23fb5e79017af0c3c874a2474078e56953e17d11","abstract_canon_sha256":"d83153da0d74d0e30ad8f81625b6b1419c5052529a5f39b15ce23fd3755bdeb9"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T00:40:59.487497Z","signature_b64":"OQbK5e9UnZb2Yfj9XHILT2jZiD57l0cD4zDVXqFOvKUupU2LWbgu79+YGZDfpcQkKr83kgAm7GWwBZeQKDjfBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ed015a239dc8eed2a631ffd9bb3fbe59b86c840ba623558a426e794b24561757","last_reissued_at":"2026-07-05T00:40:59.487013Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T00:40:59.487013Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"The Local Elasticity of Neural Networks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CV","stat.ML"],"primary_cat":"cs.LG","authors_text":"Hangfeng He, Weijie J. Su","submitted_at":"2019-10-15T17:35:30Z","abstract_excerpt":"This paper presents a phenomenon in neural networks that we refer to as \\textit{local elasticity}. Roughly speaking, a classifier is said to be locally elastic if its prediction at a feature vector $\\bx'$ is \\textit{not} significantly perturbed, after the classifier is updated via stochastic gradient descent at a (labeled) feature vector $\\bx$ that is \\textit{dissimilar} to $\\bx'$ in a certain sense. This phenomenon is shown to persist for neural networks with nonlinear activation functions through extensive simulations on real-life and synthetic datasets, whereas this is not observed in linea"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1910.06943","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":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/1910.06943/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":"1910.06943","created_at":"2026-07-05T00:40:59.487069+00:00"},{"alias_kind":"arxiv_version","alias_value":"1910.06943v2","created_at":"2026-07-05T00:40:59.487069+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1910.06943","created_at":"2026-07-05T00:40:59.487069+00:00"},{"alias_kind":"pith_short_12","alias_value":"5UAVUI45ZDXN","created_at":"2026-07-05T00:40:59.487069+00:00"},{"alias_kind":"pith_short_16","alias_value":"5UAVUI45ZDXNFJRR","created_at":"2026-07-05T00:40:59.487069+00:00"},{"alias_kind":"pith_short_8","alias_value":"5UAVUI45","created_at":"2026-07-05T00:40:59.487069+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2411.16237","citing_title":"Corrected Support Vector Regression for intraday point forecasting of prices in the continuous power market","ref_index":19,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/5UAVUI45ZDXNFJRR77M3WP56LG","json":"https://pith.science/pith/5UAVUI45ZDXNFJRR77M3WP56LG.json","graph_json":"https://pith.science/api/pith-number/5UAVUI45ZDXNFJRR77M3WP56LG/graph.json","events_json":"https://pith.science/api/pith-number/5UAVUI45ZDXNFJRR77M3WP56LG/events.json","paper":"https://pith.science/paper/5UAVUI45"},"agent_actions":{"view_html":"https://pith.science/pith/5UAVUI45ZDXNFJRR77M3WP56LG","download_json":"https://pith.science/pith/5UAVUI45ZDXNFJRR77M3WP56LG.json","view_paper":"https://pith.science/paper/5UAVUI45","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1910.06943&json=true","fetch_graph":"https://pith.science/api/pith-number/5UAVUI45ZDXNFJRR77M3WP56LG/graph.json","fetch_events":"https://pith.science/api/pith-number/5UAVUI45ZDXNFJRR77M3WP56LG/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/5UAVUI45ZDXNFJRR77M3WP56LG/action/timestamp_anchor","attest_storage":"https://pith.science/pith/5UAVUI45ZDXNFJRR77M3WP56LG/action/storage_attestation","attest_author":"https://pith.science/pith/5UAVUI45ZDXNFJRR77M3WP56LG/action/author_attestation","sign_citation":"https://pith.science/pith/5UAVUI45ZDXNFJRR77M3WP56LG/action/citation_signature","submit_replication":"https://pith.science/pith/5UAVUI45ZDXNFJRR77M3WP56LG/action/replication_record"}},"created_at":"2026-07-05T00:40:59.487069+00:00","updated_at":"2026-07-05T00:40:59.487069+00:00"}