{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2018:ETASN5KV7XTWPSNEXNKKHJIJUG","short_pith_number":"pith:ETASN5KV","canonical_record":{"source":{"id":"1809.04146","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"math.OC","submitted_at":"2018-09-11T20:37:13Z","cross_cats_sorted":[],"title_canon_sha256":"172cbec1e8ac2408ee95bb967de59e5e457c7a0b8b6654fc5580a0b9a74badb5","abstract_canon_sha256":"f8f06ff4eb9ca7ce9510c5293efea5d3e61396e1b64f2117913b66281468e4d6"},"schema_version":"1.0"},"canonical_sha256":"24c126f555fde767c9a4bb54a3a509a1998464e4232421d8153c266f33e93410","source":{"kind":"arxiv","id":"1809.04146","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1809.04146","created_at":"2026-05-17T23:55:03Z"},{"alias_kind":"arxiv_version","alias_value":"1809.04146v2","created_at":"2026-05-17T23:55:03Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1809.04146","created_at":"2026-05-17T23:55:03Z"},{"alias_kind":"pith_short_12","alias_value":"ETASN5KV7XTW","created_at":"2026-05-18T12:32:22Z"},{"alias_kind":"pith_short_16","alias_value":"ETASN5KV7XTWPSNE","created_at":"2026-05-18T12:32:22Z"},{"alias_kind":"pith_short_8","alias_value":"ETASN5KV","created_at":"2026-05-18T12:32:22Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2018:ETASN5KV7XTWPSNEXNKKHJIJUG","target":"record","payload":{"canonical_record":{"source":{"id":"1809.04146","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"math.OC","submitted_at":"2018-09-11T20:37:13Z","cross_cats_sorted":[],"title_canon_sha256":"172cbec1e8ac2408ee95bb967de59e5e457c7a0b8b6654fc5580a0b9a74badb5","abstract_canon_sha256":"f8f06ff4eb9ca7ce9510c5293efea5d3e61396e1b64f2117913b66281468e4d6"},"schema_version":"1.0"},"canonical_sha256":"24c126f555fde767c9a4bb54a3a509a1998464e4232421d8153c266f33e93410","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:55:03.318703Z","signature_b64":"uPXOaynChHm3AjOEVqnbTQ1ukgpu6I+GC6NQI1oshsYHpb1MqUzO7+ghoPIkPT4IPtNTDos2c+UR9q/nwyoeDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"24c126f555fde767c9a4bb54a3a509a1998464e4232421d8153c266f33e93410","last_reissued_at":"2026-05-17T23:55:03.318128Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:55:03.318128Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"1809.04146","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-17T23:55:03Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"1IY0IMZyInXKUYrHFZbKrvxniwIXOGPUE07C/s2AZQ9l+S7lSBw05oBq6Q1oFgzr4OWTDSbnsOPW43g+/+LsAA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-24T00:33:53.973286Z"},"content_sha256":"d63872200e5f203e1f6952a264683c568358ac7a804bd58b865313ee8323d6e6","schema_version":"1.0","event_id":"sha256:d63872200e5f203e1f6952a264683c568358ac7a804bd58b865313ee8323d6e6"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2018:ETASN5KV7XTWPSNEXNKKHJIJUG","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Nonconvex Variance Reduced Optimization with Arbitrary Sampling","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"math.OC","authors_text":"Peter Richt\\'arik, Samuel Horv\\'ath","submitted_at":"2018-09-11T20:37:13Z","abstract_excerpt":"We provide the first importance sampling variants of variance reduced algorithms for empirical risk minimization with non-convex loss functions. In particular, we analyze non-convex versions of SVRG, SAGA and SARAH. Our methods have the capacity to speed up the training process by an order of magnitude compared to the state of the art on real datasets. Moreover, we also improve upon current mini-batch analysis of these methods by proposing importance sampling for minibatches in this setting. Surprisingly, our approach can in some regimes lead to superlinear speedup with respect to the minibatc"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1809.04146","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":""},"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":"2026-05-17T23:55:03Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"kxZQVCXkKRPJYoF9ZBYkH79RJrWlpIhyYqxfIwZxaI3PaZtcaGRq975KDtGb/9nuLUWyoFCaoQKf55rlsyCeAg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-24T00:33:53.973981Z"},"content_sha256":"8d86fe44e529ebd83b4b484606539e32b0d40e03fefe8f357b142e4817de070d","schema_version":"1.0","event_id":"sha256:8d86fe44e529ebd83b4b484606539e32b0d40e03fefe8f357b142e4817de070d"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/ETASN5KV7XTWPSNEXNKKHJIJUG/bundle.json","state_url":"https://pith.science/pith/ETASN5KV7XTWPSNEXNKKHJIJUG/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/ETASN5KV7XTWPSNEXNKKHJIJUG/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":"2026-05-24T00:33:53Z","links":{"resolver":"https://pith.science/pith/ETASN5KV7XTWPSNEXNKKHJIJUG","bundle":"https://pith.science/pith/ETASN5KV7XTWPSNEXNKKHJIJUG/bundle.json","state":"https://pith.science/pith/ETASN5KV7XTWPSNEXNKKHJIJUG/state.json","well_known_bundle":"https://pith.science/.well-known/pith/ETASN5KV7XTWPSNEXNKKHJIJUG/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2018:ETASN5KV7XTWPSNEXNKKHJIJUG","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":"f8f06ff4eb9ca7ce9510c5293efea5d3e61396e1b64f2117913b66281468e4d6","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"math.OC","submitted_at":"2018-09-11T20:37:13Z","title_canon_sha256":"172cbec1e8ac2408ee95bb967de59e5e457c7a0b8b6654fc5580a0b9a74badb5"},"schema_version":"1.0","source":{"id":"1809.04146","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1809.04146","created_at":"2026-05-17T23:55:03Z"},{"alias_kind":"arxiv_version","alias_value":"1809.04146v2","created_at":"2026-05-17T23:55:03Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1809.04146","created_at":"2026-05-17T23:55:03Z"},{"alias_kind":"pith_short_12","alias_value":"ETASN5KV7XTW","created_at":"2026-05-18T12:32:22Z"},{"alias_kind":"pith_short_16","alias_value":"ETASN5KV7XTWPSNE","created_at":"2026-05-18T12:32:22Z"},{"alias_kind":"pith_short_8","alias_value":"ETASN5KV","created_at":"2026-05-18T12:32:22Z"}],"graph_snapshots":[{"event_id":"sha256:8d86fe44e529ebd83b4b484606539e32b0d40e03fefe8f357b142e4817de070d","target":"graph","created_at":"2026-05-17T23:55:03Z","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"},"paper":{"abstract_excerpt":"We provide the first importance sampling variants of variance reduced algorithms for empirical risk minimization with non-convex loss functions. In particular, we analyze non-convex versions of SVRG, SAGA and SARAH. Our methods have the capacity to speed up the training process by an order of magnitude compared to the state of the art on real datasets. Moreover, we also improve upon current mini-batch analysis of these methods by proposing importance sampling for minibatches in this setting. Surprisingly, our approach can in some regimes lead to superlinear speedup with respect to the minibatc","authors_text":"Peter Richt\\'arik, Samuel Horv\\'ath","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"math.OC","submitted_at":"2018-09-11T20:37:13Z","title":"Nonconvex Variance Reduced Optimization with Arbitrary Sampling"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1809.04146","kind":"arxiv","version":2},"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:d63872200e5f203e1f6952a264683c568358ac7a804bd58b865313ee8323d6e6","target":"record","created_at":"2026-05-17T23:55:03Z","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":"f8f06ff4eb9ca7ce9510c5293efea5d3e61396e1b64f2117913b66281468e4d6","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"math.OC","submitted_at":"2018-09-11T20:37:13Z","title_canon_sha256":"172cbec1e8ac2408ee95bb967de59e5e457c7a0b8b6654fc5580a0b9a74badb5"},"schema_version":"1.0","source":{"id":"1809.04146","kind":"arxiv","version":2}},"canonical_sha256":"24c126f555fde767c9a4bb54a3a509a1998464e4232421d8153c266f33e93410","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"24c126f555fde767c9a4bb54a3a509a1998464e4232421d8153c266f33e93410","first_computed_at":"2026-05-17T23:55:03.318128Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-17T23:55:03.318128Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"uPXOaynChHm3AjOEVqnbTQ1ukgpu6I+GC6NQI1oshsYHpb1MqUzO7+ghoPIkPT4IPtNTDos2c+UR9q/nwyoeDQ==","signature_status":"signed_v1","signed_at":"2026-05-17T23:55:03.318703Z","signed_message":"canonical_sha256_bytes"},"source_id":"1809.04146","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:d63872200e5f203e1f6952a264683c568358ac7a804bd58b865313ee8323d6e6","sha256:8d86fe44e529ebd83b4b484606539e32b0d40e03fefe8f357b142e4817de070d"],"state_sha256":"5de21b57f17840c7663098be16142145ce06ed11abffce4167d87db50a793f53"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"gfK5fPoiurFf13BaVQrXvpPgQZbNJlwtmj48oI1L4mvIMaOoRTfpdGh+3nL/+bpFG8ZDKvJS/ikmSWIXLpvgAQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-24T00:33:53.978130Z","bundle_sha256":"f44fe5c4c9054bb94b87e6fdbfd8d9f87fe984a6f2f88707c11cc57f619ed8f2"}}