{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2018:3BXVDTGTH7AUYQUC3RS2BOVBSG","short_pith_number":"pith:3BXVDTGT","canonical_record":{"source":{"id":"1803.00942","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-03-02T16:40:43Z","cross_cats_sorted":[],"title_canon_sha256":"b666b1a5ae144f295ee316724759f929019370530b4cf612592adb3c76b6abf7","abstract_canon_sha256":"aa49363bcd8b40cdd4e1870e1a21a402f11b8238445d4defc668fe7727bbef69"},"schema_version":"1.0"},"canonical_sha256":"d86f51ccd33fc14c4282dc65a0baa191bb53e1f76bea9b1c08a7dfd9f6e3bbf1","source":{"kind":"arxiv","id":"1803.00942","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1803.00942","created_at":"2026-07-05T00:14:59Z"},{"alias_kind":"arxiv_version","alias_value":"1803.00942v3","created_at":"2026-07-05T00:14:59Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1803.00942","created_at":"2026-07-05T00:14:59Z"},{"alias_kind":"pith_short_12","alias_value":"3BXVDTGTH7AU","created_at":"2026-07-05T00:14:59Z"},{"alias_kind":"pith_short_16","alias_value":"3BXVDTGTH7AUYQUC","created_at":"2026-07-05T00:14:59Z"},{"alias_kind":"pith_short_8","alias_value":"3BXVDTGT","created_at":"2026-07-05T00:14:59Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2018:3BXVDTGTH7AUYQUC3RS2BOVBSG","target":"record","payload":{"canonical_record":{"source":{"id":"1803.00942","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-03-02T16:40:43Z","cross_cats_sorted":[],"title_canon_sha256":"b666b1a5ae144f295ee316724759f929019370530b4cf612592adb3c76b6abf7","abstract_canon_sha256":"aa49363bcd8b40cdd4e1870e1a21a402f11b8238445d4defc668fe7727bbef69"},"schema_version":"1.0"},"canonical_sha256":"d86f51ccd33fc14c4282dc65a0baa191bb53e1f76bea9b1c08a7dfd9f6e3bbf1","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T00:14:59.607424Z","signature_b64":"ojw2LaBVDDK0oNY3IvPjroz11WufeFU5MlNnu7SX8h+YlctPFDEVIE/ZuoFAFjfbgQ0NcMfaxjht6RBAAHbPDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d86f51ccd33fc14c4282dc65a0baa191bb53e1f76bea9b1c08a7dfd9f6e3bbf1","last_reissued_at":"2026-07-05T00:14:59.606959Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T00:14:59.606959Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"1803.00942","source_version":3,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T00:14:59Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"dJq9/ieCehs7YbiwT5nqfJGU3qmkNcM1K/YP/mSJpa639MZb+ZprpwqafrgvgT1b7w8Q5h8X5KDLuTg4k9/MAA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-15T14:58:44.464986Z"},"content_sha256":"47b08b8ffef9dae03c11cca964eb08649f4f6d3759618389b591553a70fdda77","schema_version":"1.0","event_id":"sha256:47b08b8ffef9dae03c11cca964eb08649f4f6d3759618389b591553a70fdda77"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2018:3BXVDTGTH7AUYQUC3RS2BOVBSG","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Not All Samples Are Created Equal: Deep Learning with Importance Sampling","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Angelos Katharopoulos, Fran\\c{c}ois Fleuret","submitted_at":"2018-03-02T16:40:43Z","abstract_excerpt":"Deep neural network training spends most of the computation on examples that are properly handled, and could be ignored. We propose to mitigate this phenomenon with a principled importance sampling scheme that focuses computation on \"informative\" examples, and reduces the variance of the stochastic gradients during training. Our contribution is twofold: first, we derive a tractable upper bound to the per-sample gradient norm, and second we derive an estimator of the variance reduction achieved with importance sampling, which enables us to switch it on when it will result in an actual speedup. "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1803.00942","kind":"arxiv","version":3},"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/1803.00942/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":"2026-07-05T00:14:59Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"jhhvVGmzPQ5nXQsN73CmEsC1Y1UHahzEnJq12L7s23jDndfZUImjR6iKw0CnYFWqoDMjYGNV/EbEPpuqpC0OBg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-15T14:58:44.465482Z"},"content_sha256":"1fe97658c4dc231032f40ffe29f96bfe249f1258835f2fbd9b9b673e7e1ebe8d","schema_version":"1.0","event_id":"sha256:1fe97658c4dc231032f40ffe29f96bfe249f1258835f2fbd9b9b673e7e1ebe8d"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/3BXVDTGTH7AUYQUC3RS2BOVBSG/bundle.json","state_url":"https://pith.science/pith/3BXVDTGTH7AUYQUC3RS2BOVBSG/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/3BXVDTGTH7AUYQUC3RS2BOVBSG/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":"2026-08-15T14:58:44Z","links":{"resolver":"https://pith.science/pith/3BXVDTGTH7AUYQUC3RS2BOVBSG","bundle":"https://pith.science/pith/3BXVDTGTH7AUYQUC3RS2BOVBSG/bundle.json","state":"https://pith.science/pith/3BXVDTGTH7AUYQUC3RS2BOVBSG/state.json","well_known_bundle":"https://pith.science/.well-known/pith/3BXVDTGTH7AUYQUC3RS2BOVBSG/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2018:3BXVDTGTH7AUYQUC3RS2BOVBSG","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":"aa49363bcd8b40cdd4e1870e1a21a402f11b8238445d4defc668fe7727bbef69","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-03-02T16:40:43Z","title_canon_sha256":"b666b1a5ae144f295ee316724759f929019370530b4cf612592adb3c76b6abf7"},"schema_version":"1.0","source":{"id":"1803.00942","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1803.00942","created_at":"2026-07-05T00:14:59Z"},{"alias_kind":"arxiv_version","alias_value":"1803.00942v3","created_at":"2026-07-05T00:14:59Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1803.00942","created_at":"2026-07-05T00:14:59Z"},{"alias_kind":"pith_short_12","alias_value":"3BXVDTGTH7AU","created_at":"2026-07-05T00:14:59Z"},{"alias_kind":"pith_short_16","alias_value":"3BXVDTGTH7AUYQUC","created_at":"2026-07-05T00:14:59Z"},{"alias_kind":"pith_short_8","alias_value":"3BXVDTGT","created_at":"2026-07-05T00:14:59Z"}],"graph_snapshots":[{"event_id":"sha256:1fe97658c4dc231032f40ffe29f96bfe249f1258835f2fbd9b9b673e7e1ebe8d","target":"graph","created_at":"2026-07-05T00:14:59Z","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/1803.00942/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Deep neural network training spends most of the computation on examples that are properly handled, and could be ignored. We propose to mitigate this phenomenon with a principled importance sampling scheme that focuses computation on \"informative\" examples, and reduces the variance of the stochastic gradients during training. Our contribution is twofold: first, we derive a tractable upper bound to the per-sample gradient norm, and second we derive an estimator of the variance reduction achieved with importance sampling, which enables us to switch it on when it will result in an actual speedup. ","authors_text":"Angelos Katharopoulos, Fran\\c{c}ois Fleuret","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-03-02T16:40:43Z","title":"Not All Samples Are Created Equal: Deep Learning with Importance Sampling"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1803.00942","kind":"arxiv","version":3},"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":"sha256:47b08b8ffef9dae03c11cca964eb08649f4f6d3759618389b591553a70fdda77","target":"record","created_at":"2026-07-05T00:14:59Z","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":"aa49363bcd8b40cdd4e1870e1a21a402f11b8238445d4defc668fe7727bbef69","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-03-02T16:40:43Z","title_canon_sha256":"b666b1a5ae144f295ee316724759f929019370530b4cf612592adb3c76b6abf7"},"schema_version":"1.0","source":{"id":"1803.00942","kind":"arxiv","version":3}},"canonical_sha256":"d86f51ccd33fc14c4282dc65a0baa191bb53e1f76bea9b1c08a7dfd9f6e3bbf1","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"d86f51ccd33fc14c4282dc65a0baa191bb53e1f76bea9b1c08a7dfd9f6e3bbf1","first_computed_at":"2026-07-05T00:14:59.606959Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T00:14:59.606959Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"ojw2LaBVDDK0oNY3IvPjroz11WufeFU5MlNnu7SX8h+YlctPFDEVIE/ZuoFAFjfbgQ0NcMfaxjht6RBAAHbPDw==","signature_status":"signed_v1","signed_at":"2026-07-05T00:14:59.607424Z","signed_message":"canonical_sha256_bytes"},"source_id":"1803.00942","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:47b08b8ffef9dae03c11cca964eb08649f4f6d3759618389b591553a70fdda77","sha256:1fe97658c4dc231032f40ffe29f96bfe249f1258835f2fbd9b9b673e7e1ebe8d"],"state_sha256":"e5bbfd8dae2382bd38010cc7d2216cdad2761d4f533b8f2caa7e01acef2e848a"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"JC6M5QfVqcpq8ph6kZm6WTnnETpPYGAFY4BiODJnkxcFOq2mlPaEWktjrxNV6II8xhZazMRUbh1XtBFuBLd0Bg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-15T14:58:44.468602Z","bundle_sha256":"733b24380d300cd74fadd4600e9f1f5e8cfbf5c2016605fe3df92dc5627faaa5"}}