{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:Y53HFSELBXU5WODC6YZDPP35AQ","short_pith_number":"pith:Y53HFSEL","schema_version":"1.0","canonical_sha256":"c77672c88b0de9db3862f63237bf7d041d5ab5acf4e89481d49d0c092f7b5a0c","source":{"kind":"arxiv","id":"2308.06053","version":4},"attestation_state":"computed","paper":{"title":"Cost-effective On-device Continual Learning over Memory Hierarchy with Miro","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.AR"],"primary_cat":"cs.LG","authors_text":"Di Wang, Jonghyun Choi, Minjia Zhang, Myeongjae Jeon, Suyeon Jeong, Xinyue Ma","submitted_at":"2023-08-11T10:05:53Z","abstract_excerpt":"Continual learning (CL) trains NN models incrementally from a continuous stream of tasks. To remember previously learned knowledge, prior studies store old samples over a memory hierarchy and replay them when new tasks arrive. Edge devices that adopt CL to preserve data privacy are typically energy-sensitive and thus require high model accuracy while not compromising energy efficiency, i.e., cost-effectiveness. Our work is the first to explore the design space of hierarchical memory replay-based CL to gain insights into achieving cost-effectiveness on edge devices. We present Miro, a novel sys"},"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":"2308.06053","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2023-08-11T10:05:53Z","cross_cats_sorted":["cs.AI","cs.AR"],"title_canon_sha256":"25f62d074ed7e63f8b2db4ca0f21b3e42008c3e7bd2573b7df037e5620ee35da","abstract_canon_sha256":"f4fc62226f8bdc82f1bbe538704b9315ccc35c79a9b1c19ffb1d839aabefcf52"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:20:10.090428Z","signature_b64":"KCkC6SvqVopL/3XdCFCTtyVKTXe0lEf5hl+bZMITrZEMd/U+Skluxqo6rycTf3/J1AEXaaXfApZ5ObcVYgxgAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c77672c88b0de9db3862f63237bf7d041d5ab5acf4e89481d49d0c092f7b5a0c","last_reissued_at":"2026-07-05T07:20:10.090009Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:20:10.090009Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Cost-effective On-device Continual Learning over Memory Hierarchy with Miro","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.AR"],"primary_cat":"cs.LG","authors_text":"Di Wang, Jonghyun Choi, Minjia Zhang, Myeongjae Jeon, Suyeon Jeong, Xinyue Ma","submitted_at":"2023-08-11T10:05:53Z","abstract_excerpt":"Continual learning (CL) trains NN models incrementally from a continuous stream of tasks. To remember previously learned knowledge, prior studies store old samples over a memory hierarchy and replay them when new tasks arrive. Edge devices that adopt CL to preserve data privacy are typically energy-sensitive and thus require high model accuracy while not compromising energy efficiency, i.e., cost-effectiveness. Our work is the first to explore the design space of hierarchical memory replay-based CL to gain insights into achieving cost-effectiveness on edge devices. We present Miro, a novel sys"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2308.06053","kind":"arxiv","version":4},"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/2308.06053/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":"2308.06053","created_at":"2026-07-05T07:20:10.090065+00:00"},{"alias_kind":"arxiv_version","alias_value":"2308.06053v4","created_at":"2026-07-05T07:20:10.090065+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2308.06053","created_at":"2026-07-05T07:20:10.090065+00:00"},{"alias_kind":"pith_short_12","alias_value":"Y53HFSELBXU5","created_at":"2026-07-05T07:20:10.090065+00:00"},{"alias_kind":"pith_short_16","alias_value":"Y53HFSELBXU5WODC","created_at":"2026-07-05T07:20:10.090065+00:00"},{"alias_kind":"pith_short_8","alias_value":"Y53HFSEL","created_at":"2026-07-05T07:20:10.090065+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/Y53HFSELBXU5WODC6YZDPP35AQ","json":"https://pith.science/pith/Y53HFSELBXU5WODC6YZDPP35AQ.json","graph_json":"https://pith.science/api/pith-number/Y53HFSELBXU5WODC6YZDPP35AQ/graph.json","events_json":"https://pith.science/api/pith-number/Y53HFSELBXU5WODC6YZDPP35AQ/events.json","paper":"https://pith.science/paper/Y53HFSEL"},"agent_actions":{"view_html":"https://pith.science/pith/Y53HFSELBXU5WODC6YZDPP35AQ","download_json":"https://pith.science/pith/Y53HFSELBXU5WODC6YZDPP35AQ.json","view_paper":"https://pith.science/paper/Y53HFSEL","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2308.06053&json=true","fetch_graph":"https://pith.science/api/pith-number/Y53HFSELBXU5WODC6YZDPP35AQ/graph.json","fetch_events":"https://pith.science/api/pith-number/Y53HFSELBXU5WODC6YZDPP35AQ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/Y53HFSELBXU5WODC6YZDPP35AQ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/Y53HFSELBXU5WODC6YZDPP35AQ/action/storage_attestation","attest_author":"https://pith.science/pith/Y53HFSELBXU5WODC6YZDPP35AQ/action/author_attestation","sign_citation":"https://pith.science/pith/Y53HFSELBXU5WODC6YZDPP35AQ/action/citation_signature","submit_replication":"https://pith.science/pith/Y53HFSELBXU5WODC6YZDPP35AQ/action/replication_record"}},"created_at":"2026-07-05T07:20:10.090065+00:00","updated_at":"2026-07-05T07:20:10.090065+00:00"}