{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:2QLWCGURHNSU2KVOVJ5XZYPSUF","short_pith_number":"pith:2QLWCGUR","schema_version":"1.0","canonical_sha256":"d417611a913b654d2aaeaa7b7ce1f2a1679a45b7580e740ec05ca6971d57f0a5","source":{"kind":"arxiv","id":"2307.09988","version":2},"attestation_state":"computed","paper":{"title":"TinyTrain: Resource-Aware Task-Adaptive Sparse Training of DNNs at the Data-Scarce Edge","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CV"],"primary_cat":"cs.LG","authors_text":"Cecilia Mascolo, Jagmohan Chauhan, Nicholas D. Lane, Rui Li, Stylianos I. Venieris, Young D. Kwon","submitted_at":"2023-07-19T13:49:12Z","abstract_excerpt":"On-device training is essential for user personalisation and privacy. With the pervasiveness of IoT devices and microcontroller units (MCUs), this task becomes more challenging due to the constrained memory and compute resources, and the limited availability of labelled user data. Nonetheless, prior works neglect the data scarcity issue, require excessively long training time (e.g. a few hours), or induce substantial accuracy loss (>10%). In this paper, we propose TinyTrain, an on-device training approach that drastically reduces training time by selectively updating parts of the model and exp"},"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":"2307.09988","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2023-07-19T13:49:12Z","cross_cats_sorted":["cs.CV"],"title_canon_sha256":"355b94acb786eefdcbb7f7681251d840c227861fccf0aea524f24a86bf9be5c0","abstract_canon_sha256":"6c5a2982dd86b7fb09583008ba0d6b901136c48bd37f0231a98e5a7553bea4ee"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:29:44.427460Z","signature_b64":"CYDyChgdDnlmyYl/0KGBOwo5J4j0L2ZJSMrpsv+viVPZ0JWG5UV0ATt9JmwvFDjYyaB9QZExDFUiAtoAH9GWDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d417611a913b654d2aaeaa7b7ce1f2a1679a45b7580e740ec05ca6971d57f0a5","last_reissued_at":"2026-07-05T08:29:44.426988Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:29:44.426988Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"TinyTrain: Resource-Aware Task-Adaptive Sparse Training of DNNs at the Data-Scarce Edge","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CV"],"primary_cat":"cs.LG","authors_text":"Cecilia Mascolo, Jagmohan Chauhan, Nicholas D. Lane, Rui Li, Stylianos I. Venieris, Young D. Kwon","submitted_at":"2023-07-19T13:49:12Z","abstract_excerpt":"On-device training is essential for user personalisation and privacy. With the pervasiveness of IoT devices and microcontroller units (MCUs), this task becomes more challenging due to the constrained memory and compute resources, and the limited availability of labelled user data. Nonetheless, prior works neglect the data scarcity issue, require excessively long training time (e.g. a few hours), or induce substantial accuracy loss (>10%). In this paper, we propose TinyTrain, an on-device training approach that drastically reduces training time by selectively updating parts of the model and exp"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2307.09988","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/2307.09988/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":"2307.09988","created_at":"2026-07-05T08:29:44.427043+00:00"},{"alias_kind":"arxiv_version","alias_value":"2307.09988v2","created_at":"2026-07-05T08:29:44.427043+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2307.09988","created_at":"2026-07-05T08:29:44.427043+00:00"},{"alias_kind":"pith_short_12","alias_value":"2QLWCGURHNSU","created_at":"2026-07-05T08:29:44.427043+00:00"},{"alias_kind":"pith_short_16","alias_value":"2QLWCGURHNSU2KVO","created_at":"2026-07-05T08:29:44.427043+00:00"},{"alias_kind":"pith_short_8","alias_value":"2QLWCGUR","created_at":"2026-07-05T08:29:44.427043+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.31226","citing_title":"What changes after deployment? A survey on On-device Learning in TinyML","ref_index":44,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/2QLWCGURHNSU2KVOVJ5XZYPSUF","json":"https://pith.science/pith/2QLWCGURHNSU2KVOVJ5XZYPSUF.json","graph_json":"https://pith.science/api/pith-number/2QLWCGURHNSU2KVOVJ5XZYPSUF/graph.json","events_json":"https://pith.science/api/pith-number/2QLWCGURHNSU2KVOVJ5XZYPSUF/events.json","paper":"https://pith.science/paper/2QLWCGUR"},"agent_actions":{"view_html":"https://pith.science/pith/2QLWCGURHNSU2KVOVJ5XZYPSUF","download_json":"https://pith.science/pith/2QLWCGURHNSU2KVOVJ5XZYPSUF.json","view_paper":"https://pith.science/paper/2QLWCGUR","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2307.09988&json=true","fetch_graph":"https://pith.science/api/pith-number/2QLWCGURHNSU2KVOVJ5XZYPSUF/graph.json","fetch_events":"https://pith.science/api/pith-number/2QLWCGURHNSU2KVOVJ5XZYPSUF/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/2QLWCGURHNSU2KVOVJ5XZYPSUF/action/timestamp_anchor","attest_storage":"https://pith.science/pith/2QLWCGURHNSU2KVOVJ5XZYPSUF/action/storage_attestation","attest_author":"https://pith.science/pith/2QLWCGURHNSU2KVOVJ5XZYPSUF/action/author_attestation","sign_citation":"https://pith.science/pith/2QLWCGURHNSU2KVOVJ5XZYPSUF/action/citation_signature","submit_replication":"https://pith.science/pith/2QLWCGURHNSU2KVOVJ5XZYPSUF/action/replication_record"}},"created_at":"2026-07-05T08:29:44.427043+00:00","updated_at":"2026-07-05T08:29:44.427043+00:00"}