{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2019:3YK4DPWL7M75D7QV5US6H6OBAZ","short_pith_number":"pith:3YK4DPWL","schema_version":"1.0","canonical_sha256":"de15c1becbfb3fd1fe15ed25e3f9c10658d03899657913ce84967ed17fdbb023","source":{"kind":"arxiv","id":"1909.09148","version":2},"attestation_state":"computed","paper":{"title":"Data Augmentation Revisited: Rethinking the Distribution Gap between Clean and Augmented Data","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CV","stat.ML"],"primary_cat":"cs.LG","authors_text":"Lingxi Xie, Qi Tian, Xin Chen, Yanfeng Wang, Ya Zhang, Zhuoxun He","submitted_at":"2019-09-19T08:36:45Z","abstract_excerpt":"Data augmentation has been widely applied as an effective methodology to improve generalization in particular when training deep neural networks. Recently, researchers proposed a few intensive data augmentation techniques, which indeed improved accuracy, yet we notice that these methods augment data have also caused a considerable gap between clean and augmented data. In this paper, we revisit this problem from an analytical perspective, for which we estimate the upper-bound of expected risk using two terms, namely, empirical risk and generalization error, respectively. We develop an understan"},"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":"1909.09148","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-09-19T08:36:45Z","cross_cats_sorted":["cs.CV","stat.ML"],"title_canon_sha256":"f06f3c51800c2baaee3be82614450538a9375f5479d0f343b2dd0d47cda5293c","abstract_canon_sha256":"ed4e6bae5a668238e391e286c25afdd647117eb83f7db931fc0b88770c004443"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T00:20:55.538781Z","signature_b64":"HXUOURko903w2opWqIN8EWJ2w3uc+N2LFxtaYjxGgNY40I5UFpuNAiOYssE8b9AO1+9YxMa+YXg5gLnIZFToAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"de15c1becbfb3fd1fe15ed25e3f9c10658d03899657913ce84967ed17fdbb023","last_reissued_at":"2026-07-05T00:20:55.538258Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T00:20:55.538258Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Data Augmentation Revisited: Rethinking the Distribution Gap between Clean and Augmented Data","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CV","stat.ML"],"primary_cat":"cs.LG","authors_text":"Lingxi Xie, Qi Tian, Xin Chen, Yanfeng Wang, Ya Zhang, Zhuoxun He","submitted_at":"2019-09-19T08:36:45Z","abstract_excerpt":"Data augmentation has been widely applied as an effective methodology to improve generalization in particular when training deep neural networks. Recently, researchers proposed a few intensive data augmentation techniques, which indeed improved accuracy, yet we notice that these methods augment data have also caused a considerable gap between clean and augmented data. In this paper, we revisit this problem from an analytical perspective, for which we estimate the upper-bound of expected risk using two terms, namely, empirical risk and generalization error, respectively. We develop an understan"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1909.09148","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/1909.09148/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":"1909.09148","created_at":"2026-07-05T00:20:55.538312+00:00"},{"alias_kind":"arxiv_version","alias_value":"1909.09148v2","created_at":"2026-07-05T00:20:55.538312+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1909.09148","created_at":"2026-07-05T00:20:55.538312+00:00"},{"alias_kind":"pith_short_12","alias_value":"3YK4DPWL7M75","created_at":"2026-07-05T00:20:55.538312+00:00"},{"alias_kind":"pith_short_16","alias_value":"3YK4DPWL7M75D7QV","created_at":"2026-07-05T00:20:55.538312+00:00"},{"alias_kind":"pith_short_8","alias_value":"3YK4DPWL","created_at":"2026-07-05T00:20:55.538312+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2607.06484","citing_title":"Assessing the Operational Impact of Poisoning Attacks over Augmented 3D Point Cloud Public Datasets for Connected and Autonomous Vehicles","ref_index":14,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/3YK4DPWL7M75D7QV5US6H6OBAZ","json":"https://pith.science/pith/3YK4DPWL7M75D7QV5US6H6OBAZ.json","graph_json":"https://pith.science/api/pith-number/3YK4DPWL7M75D7QV5US6H6OBAZ/graph.json","events_json":"https://pith.science/api/pith-number/3YK4DPWL7M75D7QV5US6H6OBAZ/events.json","paper":"https://pith.science/paper/3YK4DPWL"},"agent_actions":{"view_html":"https://pith.science/pith/3YK4DPWL7M75D7QV5US6H6OBAZ","download_json":"https://pith.science/pith/3YK4DPWL7M75D7QV5US6H6OBAZ.json","view_paper":"https://pith.science/paper/3YK4DPWL","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1909.09148&json=true","fetch_graph":"https://pith.science/api/pith-number/3YK4DPWL7M75D7QV5US6H6OBAZ/graph.json","fetch_events":"https://pith.science/api/pith-number/3YK4DPWL7M75D7QV5US6H6OBAZ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/3YK4DPWL7M75D7QV5US6H6OBAZ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/3YK4DPWL7M75D7QV5US6H6OBAZ/action/storage_attestation","attest_author":"https://pith.science/pith/3YK4DPWL7M75D7QV5US6H6OBAZ/action/author_attestation","sign_citation":"https://pith.science/pith/3YK4DPWL7M75D7QV5US6H6OBAZ/action/citation_signature","submit_replication":"https://pith.science/pith/3YK4DPWL7M75D7QV5US6H6OBAZ/action/replication_record"}},"created_at":"2026-07-05T00:20:55.538312+00:00","updated_at":"2026-07-05T00:20:55.538312+00:00"}