{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2020:NUR4AD6DPZQWCAYM2NJYZGP7H2","short_pith_number":"pith:NUR4AD6D","schema_version":"1.0","canonical_sha256":"6d23c00fc37e6161030cd3538c99ff3e837206272d233602bd6c4fb8e6b41f2f","source":{"kind":"arxiv","id":"2002.09956","version":3},"attestation_state":"computed","paper":{"title":"De-randomized PAC-Bayes Margin Bounds: Applications to Non-convex and Non-smooth Predictors","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Arindam Banerjee, Tiancong Chen, Yingxue Zhou","submitted_at":"2020-02-23T17:54:07Z","abstract_excerpt":"In spite of several notable efforts, explaining the generalization of deterministic non-smooth deep nets, e.g., ReLU-nets, has remained challenging. Existing approaches for deterministic non-smooth deep nets typically need to bound the Lipschitz constant of such deep nets but such bounds are quite large, may even increase with the training set size yielding vacuous generalization bounds. In this paper, we present a new family of de-randomized PAC-Bayes margin bounds for deterministic non-convex and non-smooth predictors, e.g., ReLU-nets. Unlike PAC-Bayes, which applies to Bayesian predictors, "},"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":"2002.09956","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2020-02-23T17:54:07Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"30eb13c038d7cc70bd7da901ef17530be44b2e177888bb35b34ad5ced660c69a","abstract_canon_sha256":"9f976109dd1d80be56ccc761b1a6c684b0c0f7696077e4b7a360fba9ef87e738"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T01:51:06.100046Z","signature_b64":"MO+J5CR5Kmop6bB4KactiUc9A7Is0ltkrgHOiWe8M8usko9f1Wl5311piSN6/a1VtBzIx6FvLCVKDErWenvLBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"6d23c00fc37e6161030cd3538c99ff3e837206272d233602bd6c4fb8e6b41f2f","last_reissued_at":"2026-07-05T01:51:06.099550Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T01:51:06.099550Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"De-randomized PAC-Bayes Margin Bounds: Applications to Non-convex and Non-smooth Predictors","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Arindam Banerjee, Tiancong Chen, Yingxue Zhou","submitted_at":"2020-02-23T17:54:07Z","abstract_excerpt":"In spite of several notable efforts, explaining the generalization of deterministic non-smooth deep nets, e.g., ReLU-nets, has remained challenging. Existing approaches for deterministic non-smooth deep nets typically need to bound the Lipschitz constant of such deep nets but such bounds are quite large, may even increase with the training set size yielding vacuous generalization bounds. In this paper, we present a new family of de-randomized PAC-Bayes margin bounds for deterministic non-convex and non-smooth predictors, e.g., ReLU-nets. Unlike PAC-Bayes, which applies to Bayesian predictors, "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2002.09956","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/2002.09956/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":"2002.09956","created_at":"2026-07-05T01:51:06.099614+00:00"},{"alias_kind":"arxiv_version","alias_value":"2002.09956v3","created_at":"2026-07-05T01:51:06.099614+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2002.09956","created_at":"2026-07-05T01:51:06.099614+00:00"},{"alias_kind":"pith_short_12","alias_value":"NUR4AD6DPZQW","created_at":"2026-07-05T01:51:06.099614+00:00"},{"alias_kind":"pith_short_16","alias_value":"NUR4AD6DPZQWCAYM","created_at":"2026-07-05T01:51:06.099614+00:00"},{"alias_kind":"pith_short_8","alias_value":"NUR4AD6D","created_at":"2026-07-05T01:51:06.099614+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.19105","citing_title":"Smoothness-Based Derandomization of PAC-Bayes Bounds","ref_index":2,"is_internal_anchor":false},{"citing_arxiv_id":"2606.19105","citing_title":"Smoothness-Based Derandomization of PAC-Bayes Bounds","ref_index":2,"is_internal_anchor":false},{"citing_arxiv_id":"2605.20347","citing_title":"Symmetrization of Loss Functions for Robust Training of Neural Networks in the Presence of Noisy Labels","ref_index":6,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/NUR4AD6DPZQWCAYM2NJYZGP7H2","json":"https://pith.science/pith/NUR4AD6DPZQWCAYM2NJYZGP7H2.json","graph_json":"https://pith.science/api/pith-number/NUR4AD6DPZQWCAYM2NJYZGP7H2/graph.json","events_json":"https://pith.science/api/pith-number/NUR4AD6DPZQWCAYM2NJYZGP7H2/events.json","paper":"https://pith.science/paper/NUR4AD6D"},"agent_actions":{"view_html":"https://pith.science/pith/NUR4AD6DPZQWCAYM2NJYZGP7H2","download_json":"https://pith.science/pith/NUR4AD6DPZQWCAYM2NJYZGP7H2.json","view_paper":"https://pith.science/paper/NUR4AD6D","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2002.09956&json=true","fetch_graph":"https://pith.science/api/pith-number/NUR4AD6DPZQWCAYM2NJYZGP7H2/graph.json","fetch_events":"https://pith.science/api/pith-number/NUR4AD6DPZQWCAYM2NJYZGP7H2/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/NUR4AD6DPZQWCAYM2NJYZGP7H2/action/timestamp_anchor","attest_storage":"https://pith.science/pith/NUR4AD6DPZQWCAYM2NJYZGP7H2/action/storage_attestation","attest_author":"https://pith.science/pith/NUR4AD6DPZQWCAYM2NJYZGP7H2/action/author_attestation","sign_citation":"https://pith.science/pith/NUR4AD6DPZQWCAYM2NJYZGP7H2/action/citation_signature","submit_replication":"https://pith.science/pith/NUR4AD6DPZQWCAYM2NJYZGP7H2/action/replication_record"}},"created_at":"2026-07-05T01:51:06.099614+00:00","updated_at":"2026-07-05T01:51:06.099614+00:00"}