{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:OAZ3LS3FOE7VJSOAPZTKIURPMV","short_pith_number":"pith:OAZ3LS3F","schema_version":"1.0","canonical_sha256":"7033b5cb65713f54c9c07e66a4522f655919b3a05dacd4f45adedfd8f9832892","source":{"kind":"arxiv","id":"2202.01838","version":1},"attestation_state":"computed","paper":{"title":"Characterizing & Finding Good Data Orderings for Fast Convergence of Sequential Gradient Methods","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Amirkeivan Mohtashami, Martin Jaggi, Sebastian Stich","submitted_at":"2022-02-03T20:38:42Z","abstract_excerpt":"While SGD, which samples from the data with replacement is widely studied in theory, a variant called Random Reshuffling (RR) is more common in practice. RR iterates through random permutations of the dataset and has been shown to converge faster than SGD. When the order is chosen deterministically, a variant called incremental gradient descent (IG), the existing convergence bounds show improvement over SGD but are worse than RR. However, these bounds do not differentiate between a good and a bad ordering and hold for the worst choice of order. Meanwhile, in some cases, choosing the right orde"},"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":"2202.01838","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2022-02-03T20:38:42Z","cross_cats_sorted":[],"title_canon_sha256":"b27841d997b72180d511304cb97f457d6c7b0c93f883673f2818547d2a609e11","abstract_canon_sha256":"2860d0a30302967eca4e9399cd7cd4f1fd47317a938af547357d643d592936f1"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:54:05.637779Z","signature_b64":"o6+BFQTEK1YOsaIuPKHLCGKgbmlUai/d9n4VZHIyoYxVs7D740BVLnWO8maDwpTLn49h0n9Ohte6/ECe06sKBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7033b5cb65713f54c9c07e66a4522f655919b3a05dacd4f45adedfd8f9832892","last_reissued_at":"2026-07-05T03:54:05.637229Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:54:05.637229Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Characterizing & Finding Good Data Orderings for Fast Convergence of Sequential Gradient Methods","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Amirkeivan Mohtashami, Martin Jaggi, Sebastian Stich","submitted_at":"2022-02-03T20:38:42Z","abstract_excerpt":"While SGD, which samples from the data with replacement is widely studied in theory, a variant called Random Reshuffling (RR) is more common in practice. RR iterates through random permutations of the dataset and has been shown to converge faster than SGD. When the order is chosen deterministically, a variant called incremental gradient descent (IG), the existing convergence bounds show improvement over SGD but are worse than RR. However, these bounds do not differentiate between a good and a bad ordering and hold for the worst choice of order. Meanwhile, in some cases, choosing the right orde"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2202.01838","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":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2202.01838/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":"2202.01838","created_at":"2026-07-05T03:54:05.637308+00:00"},{"alias_kind":"arxiv_version","alias_value":"2202.01838v1","created_at":"2026-07-05T03:54:05.637308+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2202.01838","created_at":"2026-07-05T03:54:05.637308+00:00"},{"alias_kind":"pith_short_12","alias_value":"OAZ3LS3FOE7V","created_at":"2026-07-05T03:54:05.637308+00:00"},{"alias_kind":"pith_short_16","alias_value":"OAZ3LS3FOE7VJSOA","created_at":"2026-07-05T03:54:05.637308+00:00"},{"alias_kind":"pith_short_8","alias_value":"OAZ3LS3F","created_at":"2026-07-05T03:54:05.637308+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.20150","citing_title":"TideGS: Scalable Training of Over One Billion 3D Gaussian Splatting Primitives via Out-of-Core Optimization","ref_index":7,"is_internal_anchor":false},{"citing_arxiv_id":"2605.20150","citing_title":"TideGS: Scalable Training of Over One Billion 3D Gaussian Splatting Primitives via Out-of-Core Optimization","ref_index":10,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/OAZ3LS3FOE7VJSOAPZTKIURPMV","json":"https://pith.science/pith/OAZ3LS3FOE7VJSOAPZTKIURPMV.json","graph_json":"https://pith.science/api/pith-number/OAZ3LS3FOE7VJSOAPZTKIURPMV/graph.json","events_json":"https://pith.science/api/pith-number/OAZ3LS3FOE7VJSOAPZTKIURPMV/events.json","paper":"https://pith.science/paper/OAZ3LS3F"},"agent_actions":{"view_html":"https://pith.science/pith/OAZ3LS3FOE7VJSOAPZTKIURPMV","download_json":"https://pith.science/pith/OAZ3LS3FOE7VJSOAPZTKIURPMV.json","view_paper":"https://pith.science/paper/OAZ3LS3F","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2202.01838&json=true","fetch_graph":"https://pith.science/api/pith-number/OAZ3LS3FOE7VJSOAPZTKIURPMV/graph.json","fetch_events":"https://pith.science/api/pith-number/OAZ3LS3FOE7VJSOAPZTKIURPMV/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/OAZ3LS3FOE7VJSOAPZTKIURPMV/action/timestamp_anchor","attest_storage":"https://pith.science/pith/OAZ3LS3FOE7VJSOAPZTKIURPMV/action/storage_attestation","attest_author":"https://pith.science/pith/OAZ3LS3FOE7VJSOAPZTKIURPMV/action/author_attestation","sign_citation":"https://pith.science/pith/OAZ3LS3FOE7VJSOAPZTKIURPMV/action/citation_signature","submit_replication":"https://pith.science/pith/OAZ3LS3FOE7VJSOAPZTKIURPMV/action/replication_record"}},"created_at":"2026-07-05T03:54:05.637308+00:00","updated_at":"2026-07-05T03:54:05.637308+00:00"}