{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2021:HCGNJGSI2D4EARO5OHY4GBFUOQ","short_pith_number":"pith:HCGNJGSI","canonical_record":{"source":{"id":"2106.09563","kind":"arxiv","version":5},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2021-06-17T14:45:22Z","cross_cats_sorted":["cs.CV"],"title_canon_sha256":"e17a46457adcdbc0b21e1539744036cbb47637a57a76d1d1bc268a159f188566","abstract_canon_sha256":"e788ef22a4c437a07a47fd4c74d2d15b0f05c07d01e639dfd017e4950e935570"},"schema_version":"1.0"},"canonical_sha256":"388cd49a48d0f84045dd71f1c304b474342a193364c4eefe7fdfb3cb610cc176","source":{"kind":"arxiv","id":"2106.09563","version":5},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2106.09563","created_at":"2026-07-05T04:45:10Z"},{"alias_kind":"arxiv_version","alias_value":"2106.09563v5","created_at":"2026-07-05T04:45:10Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2106.09563","created_at":"2026-07-05T04:45:10Z"},{"alias_kind":"pith_short_12","alias_value":"HCGNJGSI2D4E","created_at":"2026-07-05T04:45:10Z"},{"alias_kind":"pith_short_16","alias_value":"HCGNJGSI2D4EARO5","created_at":"2026-07-05T04:45:10Z"},{"alias_kind":"pith_short_8","alias_value":"HCGNJGSI","created_at":"2026-07-05T04:45:10Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2021:HCGNJGSI2D4EARO5OHY4GBFUOQ","target":"record","payload":{"canonical_record":{"source":{"id":"2106.09563","kind":"arxiv","version":5},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2021-06-17T14:45:22Z","cross_cats_sorted":["cs.CV"],"title_canon_sha256":"e17a46457adcdbc0b21e1539744036cbb47637a57a76d1d1bc268a159f188566","abstract_canon_sha256":"e788ef22a4c437a07a47fd4c74d2d15b0f05c07d01e639dfd017e4950e935570"},"schema_version":"1.0"},"canonical_sha256":"388cd49a48d0f84045dd71f1c304b474342a193364c4eefe7fdfb3cb610cc176","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:45:10.842041Z","signature_b64":"sufTjBYGA6FA+1ZbPLgT+bZUBNoMjZ2SPGAHxuIJp+7hMbxz4+syct918vl1F1HFwjTKawd7+1mF9IRUXuJKDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"388cd49a48d0f84045dd71f1c304b474342a193364c4eefe7fdfb3cb610cc176","last_reissued_at":"2026-07-05T04:45:10.841556Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:45:10.841556Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2106.09563","source_version":5,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T04:45:10Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"fox6eAtEp3d/y53QTf5lD9e4zT5UNFlZVS3/Omc+lFIJ3MPpYGydxoi0zuiiwNZ/oGcOUSJEbpGKYb8DUtvXCQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-21T13:23:13.109487Z"},"content_sha256":"196d12a922e2f84e4c43c109670bc564763f4d47f1ebda443dcdb4c9020d1742","schema_version":"1.0","event_id":"sha256:196d12a922e2f84e4c43c109670bc564763f4d47f1ebda443dcdb4c9020d1742"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2021:HCGNJGSI2D4EARO5OHY4GBFUOQ","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"On Anytime Learning at Macroscale","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CV"],"primary_cat":"cs.LG","authors_text":"Jing Xu, Lucas Caccia, Ludovic Denoyer, Marc'Aurelio Ranzato, Myle Ott","submitted_at":"2021-06-17T14:45:22Z","abstract_excerpt":"In many practical applications of machine learning data arrives sequentially over time in large chunks. Practitioners have then to decide how to allocate their computational budget in order to obtain the best performance at any point in time. Online learning theory for convex optimization suggests that the best strategy is to use data as soon as it arrives. However, this might not be the best strategy when using deep non-linear networks, particularly when these perform multiple passes over each chunk of data rendering the overall distribution non i.i.d.. In this paper, we formalize this learni"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2106.09563","kind":"arxiv","version":5},"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/2106.09563/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T04:45:10Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"D9jt0fkIK6++lY6qXjsr29J2ojnQZ2EZKmR/Km1xV/HIShP3eJQguBJ/VdarLcaqatdDb2+ZXyNZi2Nqk4iFDg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-21T13:23:13.109992Z"},"content_sha256":"f41e8f1adaf1d3d4fbe603eaf6c0a5a195e6b60ea63f4847acb8f9e1a46bc9d6","schema_version":"1.0","event_id":"sha256:f41e8f1adaf1d3d4fbe603eaf6c0a5a195e6b60ea63f4847acb8f9e1a46bc9d6"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/HCGNJGSI2D4EARO5OHY4GBFUOQ/bundle.json","state_url":"https://pith.science/pith/HCGNJGSI2D4EARO5OHY4GBFUOQ/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/HCGNJGSI2D4EARO5OHY4GBFUOQ/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-08-21T13:23:13Z","links":{"resolver":"https://pith.science/pith/HCGNJGSI2D4EARO5OHY4GBFUOQ","bundle":"https://pith.science/pith/HCGNJGSI2D4EARO5OHY4GBFUOQ/bundle.json","state":"https://pith.science/pith/HCGNJGSI2D4EARO5OHY4GBFUOQ/state.json","well_known_bundle":"https://pith.science/.well-known/pith/HCGNJGSI2D4EARO5OHY4GBFUOQ/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2021:HCGNJGSI2D4EARO5OHY4GBFUOQ","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"e788ef22a4c437a07a47fd4c74d2d15b0f05c07d01e639dfd017e4950e935570","cross_cats_sorted":["cs.CV"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2021-06-17T14:45:22Z","title_canon_sha256":"e17a46457adcdbc0b21e1539744036cbb47637a57a76d1d1bc268a159f188566"},"schema_version":"1.0","source":{"id":"2106.09563","kind":"arxiv","version":5}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2106.09563","created_at":"2026-07-05T04:45:10Z"},{"alias_kind":"arxiv_version","alias_value":"2106.09563v5","created_at":"2026-07-05T04:45:10Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2106.09563","created_at":"2026-07-05T04:45:10Z"},{"alias_kind":"pith_short_12","alias_value":"HCGNJGSI2D4E","created_at":"2026-07-05T04:45:10Z"},{"alias_kind":"pith_short_16","alias_value":"HCGNJGSI2D4EARO5","created_at":"2026-07-05T04:45:10Z"},{"alias_kind":"pith_short_8","alias_value":"HCGNJGSI","created_at":"2026-07-05T04:45:10Z"}],"graph_snapshots":[{"event_id":"sha256:f41e8f1adaf1d3d4fbe603eaf6c0a5a195e6b60ea63f4847acb8f9e1a46bc9d6","target":"graph","created_at":"2026-07-05T04:45:10Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2106.09563/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"In many practical applications of machine learning data arrives sequentially over time in large chunks. Practitioners have then to decide how to allocate their computational budget in order to obtain the best performance at any point in time. Online learning theory for convex optimization suggests that the best strategy is to use data as soon as it arrives. However, this might not be the best strategy when using deep non-linear networks, particularly when these perform multiple passes over each chunk of data rendering the overall distribution non i.i.d.. In this paper, we formalize this learni","authors_text":"Jing Xu, Lucas Caccia, Ludovic Denoyer, Marc'Aurelio Ranzato, Myle Ott","cross_cats":["cs.CV"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2021-06-17T14:45:22Z","title":"On Anytime Learning at Macroscale"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2106.09563","kind":"arxiv","version":5},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:196d12a922e2f84e4c43c109670bc564763f4d47f1ebda443dcdb4c9020d1742","target":"record","created_at":"2026-07-05T04:45:10Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"e788ef22a4c437a07a47fd4c74d2d15b0f05c07d01e639dfd017e4950e935570","cross_cats_sorted":["cs.CV"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2021-06-17T14:45:22Z","title_canon_sha256":"e17a46457adcdbc0b21e1539744036cbb47637a57a76d1d1bc268a159f188566"},"schema_version":"1.0","source":{"id":"2106.09563","kind":"arxiv","version":5}},"canonical_sha256":"388cd49a48d0f84045dd71f1c304b474342a193364c4eefe7fdfb3cb610cc176","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"388cd49a48d0f84045dd71f1c304b474342a193364c4eefe7fdfb3cb610cc176","first_computed_at":"2026-07-05T04:45:10.841556Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T04:45:10.841556Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"sufTjBYGA6FA+1ZbPLgT+bZUBNoMjZ2SPGAHxuIJp+7hMbxz4+syct918vl1F1HFwjTKawd7+1mF9IRUXuJKDQ==","signature_status":"signed_v1","signed_at":"2026-07-05T04:45:10.842041Z","signed_message":"canonical_sha256_bytes"},"source_id":"2106.09563","source_kind":"arxiv","source_version":5}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:196d12a922e2f84e4c43c109670bc564763f4d47f1ebda443dcdb4c9020d1742","sha256:f41e8f1adaf1d3d4fbe603eaf6c0a5a195e6b60ea63f4847acb8f9e1a46bc9d6"],"state_sha256":"129c66e295d8c4e5b6ceb38c838495ac40145a4b4d31c53b7447fbc16e1d14b9"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"H/G80xPnK7Ks4PcndaYnp3p14yLJlyN1NbNrP2tlqe7Qsd7eKtM3j5OoTJvqmViley38ksmN7aTDNq1a3mxdCA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-21T13:23:13.113928Z","bundle_sha256":"d4e26458cc6ad81ae36394f19b61de2bae0b2a55a2a23cdd1bf256a912a1ef8c"}}