{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:65JI2NT53ZGTXQQP5XIHOYF5DM","short_pith_number":"pith:65JI2NT5","schema_version":"1.0","canonical_sha256":"f7528d367dde4d3bc20fedd07760bd1b30d73d85c5a91f7ae5581d668ee9fc1f","source":{"kind":"arxiv","id":"2104.04282","version":2},"attestation_state":"computed","paper":{"title":"Direct Differentiable Augmentation Search","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Aoming Liu, Naiyan Wang, Zehao Huang, Zhiwu Huang","submitted_at":"2021-04-09T10:02:24Z","abstract_excerpt":"Data augmentation has been an indispensable tool to improve the performance of deep neural networks, however the augmentation can hardly transfer among different tasks and datasets. Consequently, a recent trend is to adopt AutoML technique to learn proper augmentation policy without extensive hand-crafted tuning. In this paper, we propose an efficient differentiable search algorithm called Direct Differentiable Augmentation Search (DDAS). It exploits meta-learning with one-step gradient update and continuous relaxation to the expected training loss for efficient search. Our DDAS can achieve ef"},"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":"2104.04282","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2021-04-09T10:02:24Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"4835d5622635d15cc39fbfeceb98dc1ff0182862b33bbf1c9396065ed8b1218e","abstract_canon_sha256":"2df3630b0b42462f94aee98e9ecc2e4a3d6702da86e9559a2f5465b0a39d61bd"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:21:46.464801Z","signature_b64":"RQYilOjxJPWDeirweanSRTCRndOoMc24EP2hemBZNBLWD9QKSDG0X0z10+R/cUrY4R5J+JW6IQVZjvKh0e8ECQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f7528d367dde4d3bc20fedd07760bd1b30d73d85c5a91f7ae5581d668ee9fc1f","last_reissued_at":"2026-07-05T03:21:46.464302Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:21:46.464302Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Direct Differentiable Augmentation Search","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Aoming Liu, Naiyan Wang, Zehao Huang, Zhiwu Huang","submitted_at":"2021-04-09T10:02:24Z","abstract_excerpt":"Data augmentation has been an indispensable tool to improve the performance of deep neural networks, however the augmentation can hardly transfer among different tasks and datasets. Consequently, a recent trend is to adopt AutoML technique to learn proper augmentation policy without extensive hand-crafted tuning. In this paper, we propose an efficient differentiable search algorithm called Direct Differentiable Augmentation Search (DDAS). It exploits meta-learning with one-step gradient update and continuous relaxation to the expected training loss for efficient search. Our DDAS can achieve ef"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2104.04282","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/2104.04282/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":"2104.04282","created_at":"2026-07-05T03:21:46.464379+00:00"},{"alias_kind":"arxiv_version","alias_value":"2104.04282v2","created_at":"2026-07-05T03:21:46.464379+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2104.04282","created_at":"2026-07-05T03:21:46.464379+00:00"},{"alias_kind":"pith_short_12","alias_value":"65JI2NT53ZGT","created_at":"2026-07-05T03:21:46.464379+00:00"},{"alias_kind":"pith_short_16","alias_value":"65JI2NT53ZGTXQQP","created_at":"2026-07-05T03:21:46.464379+00:00"},{"alias_kind":"pith_short_8","alias_value":"65JI2NT5","created_at":"2026-07-05T03:21:46.464379+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/65JI2NT53ZGTXQQP5XIHOYF5DM","json":"https://pith.science/pith/65JI2NT53ZGTXQQP5XIHOYF5DM.json","graph_json":"https://pith.science/api/pith-number/65JI2NT53ZGTXQQP5XIHOYF5DM/graph.json","events_json":"https://pith.science/api/pith-number/65JI2NT53ZGTXQQP5XIHOYF5DM/events.json","paper":"https://pith.science/paper/65JI2NT5"},"agent_actions":{"view_html":"https://pith.science/pith/65JI2NT53ZGTXQQP5XIHOYF5DM","download_json":"https://pith.science/pith/65JI2NT53ZGTXQQP5XIHOYF5DM.json","view_paper":"https://pith.science/paper/65JI2NT5","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2104.04282&json=true","fetch_graph":"https://pith.science/api/pith-number/65JI2NT53ZGTXQQP5XIHOYF5DM/graph.json","fetch_events":"https://pith.science/api/pith-number/65JI2NT53ZGTXQQP5XIHOYF5DM/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/65JI2NT53ZGTXQQP5XIHOYF5DM/action/timestamp_anchor","attest_storage":"https://pith.science/pith/65JI2NT53ZGTXQQP5XIHOYF5DM/action/storage_attestation","attest_author":"https://pith.science/pith/65JI2NT53ZGTXQQP5XIHOYF5DM/action/author_attestation","sign_citation":"https://pith.science/pith/65JI2NT53ZGTXQQP5XIHOYF5DM/action/citation_signature","submit_replication":"https://pith.science/pith/65JI2NT53ZGTXQQP5XIHOYF5DM/action/replication_record"}},"created_at":"2026-07-05T03:21:46.464379+00:00","updated_at":"2026-07-05T03:21:46.464379+00:00"}