{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:SWQOD3HOF6SCYNOYCEYEYALZTX","short_pith_number":"pith:SWQOD3HO","schema_version":"1.0","canonical_sha256":"95a0e1ecee2fa42c35d811304c01799df141d2c7282316ed9a30facaa6bb6588","source":{"kind":"arxiv","id":"2311.12727","version":2},"attestation_state":"computed","paper":{"title":"Soft Random Sampling: A Theoretical and Empirical Analysis","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.LG","authors_text":"Ashish Mittal, Brian Kingsbury, George Saon, Songtao Lu, Wei Zhang, Xiaodong Cui","submitted_at":"2023-11-21T17:03:21Z","abstract_excerpt":"Soft random sampling (SRS) is a simple yet effective approach for efficient training of large-scale deep neural networks when dealing with massive data. SRS selects a subset uniformly at random with replacement from the full data set in each epoch. In this paper, we conduct a theoretical and empirical analysis of SRS. First, we analyze its sampling dynamics including data coverage and occupancy. Next, we investigate its convergence with non-convex objective functions and give the convergence rate. Finally, we provide its generalization performance. We empirically evaluate SRS for image recogni"},"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":"2311.12727","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2023-11-21T17:03:21Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"2f539123a36e43fdbb689a5cd62fc2f07a9c543230247426c636457972da8310","abstract_canon_sha256":"d421769cc9ccea79b8b2f592ebd9e9a042c7e3f93200184b6d44c3ed8fb6fbb7"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:16:04.740686Z","signature_b64":"V55U5LBqEBkVJPzZkHL4lGtHG3VWeuksFRoRWF7AiheXq5A5wq8wGmU/FFFFeM70ZLyjBr5VG1spxc4grH8HAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"95a0e1ecee2fa42c35d811304c01799df141d2c7282316ed9a30facaa6bb6588","last_reissued_at":"2026-07-05T07:16:04.740191Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:16:04.740191Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Soft Random Sampling: A Theoretical and Empirical Analysis","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.LG","authors_text":"Ashish Mittal, Brian Kingsbury, George Saon, Songtao Lu, Wei Zhang, Xiaodong Cui","submitted_at":"2023-11-21T17:03:21Z","abstract_excerpt":"Soft random sampling (SRS) is a simple yet effective approach for efficient training of large-scale deep neural networks when dealing with massive data. SRS selects a subset uniformly at random with replacement from the full data set in each epoch. In this paper, we conduct a theoretical and empirical analysis of SRS. First, we analyze its sampling dynamics including data coverage and occupancy. Next, we investigate its convergence with non-convex objective functions and give the convergence rate. Finally, we provide its generalization performance. We empirically evaluate SRS for image recogni"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2311.12727","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/2311.12727/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":"2311.12727","created_at":"2026-07-05T07:16:04.740246+00:00"},{"alias_kind":"arxiv_version","alias_value":"2311.12727v2","created_at":"2026-07-05T07:16:04.740246+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2311.12727","created_at":"2026-07-05T07:16:04.740246+00:00"},{"alias_kind":"pith_short_12","alias_value":"SWQOD3HOF6SC","created_at":"2026-07-05T07:16:04.740246+00:00"},{"alias_kind":"pith_short_16","alias_value":"SWQOD3HOF6SCYNOY","created_at":"2026-07-05T07:16:04.740246+00:00"},{"alias_kind":"pith_short_8","alias_value":"SWQOD3HO","created_at":"2026-07-05T07:16:04.740246+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2507.09753","citing_title":"Do we need equivariant models for molecule generation?","ref_index":2018,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/SWQOD3HOF6SCYNOYCEYEYALZTX","json":"https://pith.science/pith/SWQOD3HOF6SCYNOYCEYEYALZTX.json","graph_json":"https://pith.science/api/pith-number/SWQOD3HOF6SCYNOYCEYEYALZTX/graph.json","events_json":"https://pith.science/api/pith-number/SWQOD3HOF6SCYNOYCEYEYALZTX/events.json","paper":"https://pith.science/paper/SWQOD3HO"},"agent_actions":{"view_html":"https://pith.science/pith/SWQOD3HOF6SCYNOYCEYEYALZTX","download_json":"https://pith.science/pith/SWQOD3HOF6SCYNOYCEYEYALZTX.json","view_paper":"https://pith.science/paper/SWQOD3HO","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2311.12727&json=true","fetch_graph":"https://pith.science/api/pith-number/SWQOD3HOF6SCYNOYCEYEYALZTX/graph.json","fetch_events":"https://pith.science/api/pith-number/SWQOD3HOF6SCYNOYCEYEYALZTX/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/SWQOD3HOF6SCYNOYCEYEYALZTX/action/timestamp_anchor","attest_storage":"https://pith.science/pith/SWQOD3HOF6SCYNOYCEYEYALZTX/action/storage_attestation","attest_author":"https://pith.science/pith/SWQOD3HOF6SCYNOYCEYEYALZTX/action/author_attestation","sign_citation":"https://pith.science/pith/SWQOD3HOF6SCYNOYCEYEYALZTX/action/citation_signature","submit_replication":"https://pith.science/pith/SWQOD3HOF6SCYNOYCEYEYALZTX/action/replication_record"}},"created_at":"2026-07-05T07:16:04.740246+00:00","updated_at":"2026-07-05T07:16:04.740246+00:00"}