{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:LPCGO4WLPPLTHY7SPCCOTLR56B","short_pith_number":"pith:LPCGO4WL","schema_version":"1.0","canonical_sha256":"5bc46772cb7bd733e3f27884e9ae3df06f8acc4ae3b9784ad882d6015e6b1c0b","source":{"kind":"arxiv","id":"2110.08890","version":2},"attestation_state":"computed","paper":{"title":"Network Augmentation for Tiny Deep Learning","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CV","authors_text":"Chuang Gan, Han Cai, Ji Lin, Song Han","submitted_at":"2021-10-17T18:48:41Z","abstract_excerpt":"We introduce Network Augmentation (NetAug), a new training method for improving the performance of tiny neural networks. Existing regularization techniques (e.g., data augmentation, dropout) have shown much success on large neural networks by adding noise to overcome over-fitting. However, we found these techniques hurt the performance of tiny neural networks. We argue that training tiny models are different from large models: rather than augmenting the data, we should augment the model, since tiny models tend to suffer from under-fitting rather than over-fitting due to limited capacity. To al"},"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":"2110.08890","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CV","submitted_at":"2021-10-17T18:48:41Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"e31a7b4dcdf11d8cc60803be8acdb00e3b1277d434a151877043bc3d441138cb","abstract_canon_sha256":"88dd2b6a5216919764dc82d230dde6df56680859c5863f824da0e04c3a633a33"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:17:17.711697Z","signature_b64":"kTSqj38x0B/agugg15913Mz6NsedfKns7qufBddgzM8MT/ckcliKqPXt1weo2AIaiVADihGLQBJ3dCcJGQ0kBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"5bc46772cb7bd733e3f27884e9ae3df06f8acc4ae3b9784ad882d6015e6b1c0b","last_reissued_at":"2026-07-05T04:17:17.711087Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:17:17.711087Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Network Augmentation for Tiny Deep Learning","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CV","authors_text":"Chuang Gan, Han Cai, Ji Lin, Song Han","submitted_at":"2021-10-17T18:48:41Z","abstract_excerpt":"We introduce Network Augmentation (NetAug), a new training method for improving the performance of tiny neural networks. Existing regularization techniques (e.g., data augmentation, dropout) have shown much success on large neural networks by adding noise to overcome over-fitting. However, we found these techniques hurt the performance of tiny neural networks. We argue that training tiny models are different from large models: rather than augmenting the data, we should augment the model, since tiny models tend to suffer from under-fitting rather than over-fitting due to limited capacity. To al"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2110.08890","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/2110.08890/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":"2110.08890","created_at":"2026-07-05T04:17:17.711168+00:00"},{"alias_kind":"arxiv_version","alias_value":"2110.08890v2","created_at":"2026-07-05T04:17:17.711168+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2110.08890","created_at":"2026-07-05T04:17:17.711168+00:00"},{"alias_kind":"pith_short_12","alias_value":"LPCGO4WLPPLT","created_at":"2026-07-05T04:17:17.711168+00:00"},{"alias_kind":"pith_short_16","alias_value":"LPCGO4WLPPLTHY7S","created_at":"2026-07-05T04:17:17.711168+00:00"},{"alias_kind":"pith_short_8","alias_value":"LPCGO4WL","created_at":"2026-07-05T04:17:17.711168+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/LPCGO4WLPPLTHY7SPCCOTLR56B","json":"https://pith.science/pith/LPCGO4WLPPLTHY7SPCCOTLR56B.json","graph_json":"https://pith.science/api/pith-number/LPCGO4WLPPLTHY7SPCCOTLR56B/graph.json","events_json":"https://pith.science/api/pith-number/LPCGO4WLPPLTHY7SPCCOTLR56B/events.json","paper":"https://pith.science/paper/LPCGO4WL"},"agent_actions":{"view_html":"https://pith.science/pith/LPCGO4WLPPLTHY7SPCCOTLR56B","download_json":"https://pith.science/pith/LPCGO4WLPPLTHY7SPCCOTLR56B.json","view_paper":"https://pith.science/paper/LPCGO4WL","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2110.08890&json=true","fetch_graph":"https://pith.science/api/pith-number/LPCGO4WLPPLTHY7SPCCOTLR56B/graph.json","fetch_events":"https://pith.science/api/pith-number/LPCGO4WLPPLTHY7SPCCOTLR56B/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/LPCGO4WLPPLTHY7SPCCOTLR56B/action/timestamp_anchor","attest_storage":"https://pith.science/pith/LPCGO4WLPPLTHY7SPCCOTLR56B/action/storage_attestation","attest_author":"https://pith.science/pith/LPCGO4WLPPLTHY7SPCCOTLR56B/action/author_attestation","sign_citation":"https://pith.science/pith/LPCGO4WLPPLTHY7SPCCOTLR56B/action/citation_signature","submit_replication":"https://pith.science/pith/LPCGO4WLPPLTHY7SPCCOTLR56B/action/replication_record"}},"created_at":"2026-07-05T04:17:17.711168+00:00","updated_at":"2026-07-05T04:17:17.711168+00:00"}