{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:BOVLCZRSUHJML66VE4WSYU3AJC","short_pith_number":"pith:BOVLCZRS","schema_version":"1.0","canonical_sha256":"0baab16632a1d2c5fbd5272d2c536048b02c8cbf422ac2c8da06a0e7bfa10435","source":{"kind":"arxiv","id":"2208.08567","version":2},"attestation_state":"computed","paper":{"title":"High Probability Bounds for Stochastic Subgradient Schemes with Heavy Tailed Noise","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"math.OC","authors_text":"Andrea Paudice, Daniela A. Parletta, Massimiliano Pontil, Saverio Salzo","submitted_at":"2022-08-17T23:05:05Z","abstract_excerpt":"In this work we study high probability bounds for stochastic subgradient methods under heavy tailed noise. In this setting the noise is only assumed to have finite variance as opposed to a sub-Gaussian distribution for which it is known that standard subgradient methods enjoys high probability bounds. We analyzed a clipped version of the projected stochastic subgradient method, where subgradient estimates are truncated whenever they have large norms. We show that this clipping strategy leads both to near optimal any-time and finite horizon bounds for many classical averaging schemes. Prelimina"},"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":"2208.08567","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"math.OC","submitted_at":"2022-08-17T23:05:05Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"a234c61120abd3dc1af316cf47079c5791cc44a306d5020ce2c20c510106cddb","abstract_canon_sha256":"80013ebf39b8a0d69a6f65063a4c15d9c111089a6f979a14c32baf51d88978ac"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:07:54.434252Z","signature_b64":"CEu3Ipoji5vcNUhu2GiAO0goya0amR6c38FYMYGjBA1IxfXLF9s7I9ZtbiBRO/+hstT6lKtWHWc84yoy5ZM4DQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0baab16632a1d2c5fbd5272d2c536048b02c8cbf422ac2c8da06a0e7bfa10435","last_reissued_at":"2026-07-05T08:07:54.433791Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:07:54.433791Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"High Probability Bounds for Stochastic Subgradient Schemes with Heavy Tailed Noise","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"math.OC","authors_text":"Andrea Paudice, Daniela A. Parletta, Massimiliano Pontil, Saverio Salzo","submitted_at":"2022-08-17T23:05:05Z","abstract_excerpt":"In this work we study high probability bounds for stochastic subgradient methods under heavy tailed noise. In this setting the noise is only assumed to have finite variance as opposed to a sub-Gaussian distribution for which it is known that standard subgradient methods enjoys high probability bounds. We analyzed a clipped version of the projected stochastic subgradient method, where subgradient estimates are truncated whenever they have large norms. We show that this clipping strategy leads both to near optimal any-time and finite horizon bounds for many classical averaging schemes. Prelimina"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2208.08567","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/2208.08567/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":"2208.08567","created_at":"2026-07-05T08:07:54.433851+00:00"},{"alias_kind":"arxiv_version","alias_value":"2208.08567v2","created_at":"2026-07-05T08:07:54.433851+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2208.08567","created_at":"2026-07-05T08:07:54.433851+00:00"},{"alias_kind":"pith_short_12","alias_value":"BOVLCZRSUHJM","created_at":"2026-07-05T08:07:54.433851+00:00"},{"alias_kind":"pith_short_16","alias_value":"BOVLCZRSUHJML66V","created_at":"2026-07-05T08:07:54.433851+00:00"},{"alias_kind":"pith_short_8","alias_value":"BOVLCZRS","created_at":"2026-07-05T08:07:54.433851+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/BOVLCZRSUHJML66VE4WSYU3AJC","json":"https://pith.science/pith/BOVLCZRSUHJML66VE4WSYU3AJC.json","graph_json":"https://pith.science/api/pith-number/BOVLCZRSUHJML66VE4WSYU3AJC/graph.json","events_json":"https://pith.science/api/pith-number/BOVLCZRSUHJML66VE4WSYU3AJC/events.json","paper":"https://pith.science/paper/BOVLCZRS"},"agent_actions":{"view_html":"https://pith.science/pith/BOVLCZRSUHJML66VE4WSYU3AJC","download_json":"https://pith.science/pith/BOVLCZRSUHJML66VE4WSYU3AJC.json","view_paper":"https://pith.science/paper/BOVLCZRS","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2208.08567&json=true","fetch_graph":"https://pith.science/api/pith-number/BOVLCZRSUHJML66VE4WSYU3AJC/graph.json","fetch_events":"https://pith.science/api/pith-number/BOVLCZRSUHJML66VE4WSYU3AJC/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/BOVLCZRSUHJML66VE4WSYU3AJC/action/timestamp_anchor","attest_storage":"https://pith.science/pith/BOVLCZRSUHJML66VE4WSYU3AJC/action/storage_attestation","attest_author":"https://pith.science/pith/BOVLCZRSUHJML66VE4WSYU3AJC/action/author_attestation","sign_citation":"https://pith.science/pith/BOVLCZRSUHJML66VE4WSYU3AJC/action/citation_signature","submit_replication":"https://pith.science/pith/BOVLCZRSUHJML66VE4WSYU3AJC/action/replication_record"}},"created_at":"2026-07-05T08:07:54.433851+00:00","updated_at":"2026-07-05T08:07:54.433851+00:00"}