{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2019:VUWZFWR64HX5WB6TRDOKXRTNY5","short_pith_number":"pith:VUWZFWR6","schema_version":"1.0","canonical_sha256":"ad2d92da3ee1efdb07d388dcabc66dc76d20728679ca8b3d6725180a15230c79","source":{"kind":"arxiv","id":"1902.02434","version":1},"attestation_state":"computed","paper":{"title":"A Scale Invariant Flatness Measure for Deep Network Minima","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"stat.ML","authors_text":"Abhishek Kumar, Akshay Rangamani, Dzung Phan, Nam H. Nguyen, Sang H. Chin, Trac D. Tran","submitted_at":"2019-02-06T23:51:53Z","abstract_excerpt":"It has been empirically observed that the flatness of minima obtained from training deep networks seems to correlate with better generalization. However, for deep networks with positively homogeneous activations, most measures of sharpness/flatness are not invariant to rescaling of the network parameters, corresponding to the same function. This means that the measure of flatness/sharpness can be made as small or as large as possible through rescaling, rendering the quantitative measures meaningless. In this paper we show that for deep networks with positively homogenous activations, these res"},"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":"1902.02434","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"stat.ML","submitted_at":"2019-02-06T23:51:53Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"060bd9f9e6405bdea801e9bd3ef0743c9cbe1238e6b1e470093d38478bab1518","abstract_canon_sha256":"8e27e653e872651f979a2a5136f89606751eabd4182892ae10bd00454047e8f3"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:54:33.335252Z","signature_b64":"4bXrWsKro6ibtOsOjj4JosFqpF8qH80AZq9mGqnd2zxJKVGutEvqy2zZIw6VM1rpNjPK8Usm2GxeW1RC7TQCAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ad2d92da3ee1efdb07d388dcabc66dc76d20728679ca8b3d6725180a15230c79","last_reissued_at":"2026-05-17T23:54:33.334586Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:54:33.334586Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"A Scale Invariant Flatness Measure for Deep Network Minima","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"stat.ML","authors_text":"Abhishek Kumar, Akshay Rangamani, Dzung Phan, Nam H. Nguyen, Sang H. Chin, Trac D. Tran","submitted_at":"2019-02-06T23:51:53Z","abstract_excerpt":"It has been empirically observed that the flatness of minima obtained from training deep networks seems to correlate with better generalization. However, for deep networks with positively homogeneous activations, most measures of sharpness/flatness are not invariant to rescaling of the network parameters, corresponding to the same function. This means that the measure of flatness/sharpness can be made as small or as large as possible through rescaling, rendering the quantitative measures meaningless. In this paper we show that for deep networks with positively homogenous activations, these res"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1902.02434","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":""},"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":"1902.02434","created_at":"2026-05-17T23:54:33.334695+00:00"},{"alias_kind":"arxiv_version","alias_value":"1902.02434v1","created_at":"2026-05-17T23:54:33.334695+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1902.02434","created_at":"2026-05-17T23:54:33.334695+00:00"},{"alias_kind":"pith_short_12","alias_value":"VUWZFWR64HX5","created_at":"2026-05-18T12:33:30.264802+00:00"},{"alias_kind":"pith_short_16","alias_value":"VUWZFWR64HX5WB6T","created_at":"2026-05-18T12:33:30.264802+00:00"},{"alias_kind":"pith_short_8","alias_value":"VUWZFWR6","created_at":"2026-05-18T12:33:30.264802+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"1907.10732","citing_title":"Hessian based analysis of SGD for Deep Nets: Dynamics and Generalization","ref_index":57,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/VUWZFWR64HX5WB6TRDOKXRTNY5","json":"https://pith.science/pith/VUWZFWR64HX5WB6TRDOKXRTNY5.json","graph_json":"https://pith.science/api/pith-number/VUWZFWR64HX5WB6TRDOKXRTNY5/graph.json","events_json":"https://pith.science/api/pith-number/VUWZFWR64HX5WB6TRDOKXRTNY5/events.json","paper":"https://pith.science/paper/VUWZFWR6"},"agent_actions":{"view_html":"https://pith.science/pith/VUWZFWR64HX5WB6TRDOKXRTNY5","download_json":"https://pith.science/pith/VUWZFWR64HX5WB6TRDOKXRTNY5.json","view_paper":"https://pith.science/paper/VUWZFWR6","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1902.02434&json=true","fetch_graph":"https://pith.science/api/pith-number/VUWZFWR64HX5WB6TRDOKXRTNY5/graph.json","fetch_events":"https://pith.science/api/pith-number/VUWZFWR64HX5WB6TRDOKXRTNY5/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/VUWZFWR64HX5WB6TRDOKXRTNY5/action/timestamp_anchor","attest_storage":"https://pith.science/pith/VUWZFWR64HX5WB6TRDOKXRTNY5/action/storage_attestation","attest_author":"https://pith.science/pith/VUWZFWR64HX5WB6TRDOKXRTNY5/action/author_attestation","sign_citation":"https://pith.science/pith/VUWZFWR64HX5WB6TRDOKXRTNY5/action/citation_signature","submit_replication":"https://pith.science/pith/VUWZFWR64HX5WB6TRDOKXRTNY5/action/replication_record"}},"created_at":"2026-05-17T23:54:33.334695+00:00","updated_at":"2026-05-17T23:54:33.334695+00:00"}