{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2020:NACLHSBPATKDAELUTURFNARFEY","short_pith_number":"pith:NACLHSBP","schema_version":"1.0","canonical_sha256":"6804b3c82f04d43011749d22568225260f9d4f8ed4fe19ea3204c9068a7b8be0","source":{"kind":"arxiv","id":"2005.10406","version":2},"attestation_state":"computed","paper":{"title":"Training Keyword Spotting Models on Non-IID Data with Federated Learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL","cs.LG","cs.SD"],"primary_cat":"eess.AS","authors_text":"Aishanee Shah, Andrew Hard, Cameron Nguyen, Ignacio Lopez Moreno, Kurt Partridge, Niranjan Subrahmanya, Pai Zhu, Rajiv Mathews","submitted_at":"2020-05-21T00:53:33Z","abstract_excerpt":"We demonstrate that a production-quality keyword-spotting model can be trained on-device using federated learning and achieve comparable false accept and false reject rates to a centrally-trained model. To overcome the algorithmic constraints associated with fitting on-device data (which are inherently non-independent and identically distributed), we conduct thorough empirical studies of optimization algorithms and hyperparameter configurations using large-scale federated simulations. To overcome resource constraints, we replace memory intensive MTR data augmentation with SpecAugment, which re"},"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":"2005.10406","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"eess.AS","submitted_at":"2020-05-21T00:53:33Z","cross_cats_sorted":["cs.CL","cs.LG","cs.SD"],"title_canon_sha256":"2a5e506ed6a6ed9cce67b32c87ff532e7c2f3982606657cb394d1b3cc0487b4d","abstract_canon_sha256":"db809fc7ea18d14124a01e6d46a507d3c4603986d887ede39e04a732676921ae"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T01:07:54.025027Z","signature_b64":"OGeO0pbvYRFQr6rSPMeKaPLAjKRSVQM0LWETBOzvjOYWoHn2lNnMjyIPbjKDkkNf7rZPYVbQZVcdKYmz41ExDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"6804b3c82f04d43011749d22568225260f9d4f8ed4fe19ea3204c9068a7b8be0","last_reissued_at":"2026-07-05T01:07:54.024573Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T01:07:54.024573Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Training Keyword Spotting Models on Non-IID Data with Federated Learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL","cs.LG","cs.SD"],"primary_cat":"eess.AS","authors_text":"Aishanee Shah, Andrew Hard, Cameron Nguyen, Ignacio Lopez Moreno, Kurt Partridge, Niranjan Subrahmanya, Pai Zhu, Rajiv Mathews","submitted_at":"2020-05-21T00:53:33Z","abstract_excerpt":"We demonstrate that a production-quality keyword-spotting model can be trained on-device using federated learning and achieve comparable false accept and false reject rates to a centrally-trained model. To overcome the algorithmic constraints associated with fitting on-device data (which are inherently non-independent and identically distributed), we conduct thorough empirical studies of optimization algorithms and hyperparameter configurations using large-scale federated simulations. To overcome resource constraints, we replace memory intensive MTR data augmentation with SpecAugment, which re"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2005.10406","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/2005.10406/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":"2005.10406","created_at":"2026-07-05T01:07:54.024627+00:00"},{"alias_kind":"arxiv_version","alias_value":"2005.10406v2","created_at":"2026-07-05T01:07:54.024627+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2005.10406","created_at":"2026-07-05T01:07:54.024627+00:00"},{"alias_kind":"pith_short_12","alias_value":"NACLHSBPATKD","created_at":"2026-07-05T01:07:54.024627+00:00"},{"alias_kind":"pith_short_16","alias_value":"NACLHSBPATKDAELU","created_at":"2026-07-05T01:07:54.024627+00:00"},{"alias_kind":"pith_short_8","alias_value":"NACLHSBP","created_at":"2026-07-05T01:07:54.024627+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2505.19823","citing_title":"LAPA-based Dynamic Privacy Optimization for Wireless Federated Learning in Heterogeneous Environments","ref_index":2,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/NACLHSBPATKDAELUTURFNARFEY","json":"https://pith.science/pith/NACLHSBPATKDAELUTURFNARFEY.json","graph_json":"https://pith.science/api/pith-number/NACLHSBPATKDAELUTURFNARFEY/graph.json","events_json":"https://pith.science/api/pith-number/NACLHSBPATKDAELUTURFNARFEY/events.json","paper":"https://pith.science/paper/NACLHSBP"},"agent_actions":{"view_html":"https://pith.science/pith/NACLHSBPATKDAELUTURFNARFEY","download_json":"https://pith.science/pith/NACLHSBPATKDAELUTURFNARFEY.json","view_paper":"https://pith.science/paper/NACLHSBP","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2005.10406&json=true","fetch_graph":"https://pith.science/api/pith-number/NACLHSBPATKDAELUTURFNARFEY/graph.json","fetch_events":"https://pith.science/api/pith-number/NACLHSBPATKDAELUTURFNARFEY/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/NACLHSBPATKDAELUTURFNARFEY/action/timestamp_anchor","attest_storage":"https://pith.science/pith/NACLHSBPATKDAELUTURFNARFEY/action/storage_attestation","attest_author":"https://pith.science/pith/NACLHSBPATKDAELUTURFNARFEY/action/author_attestation","sign_citation":"https://pith.science/pith/NACLHSBPATKDAELUTURFNARFEY/action/citation_signature","submit_replication":"https://pith.science/pith/NACLHSBPATKDAELUTURFNARFEY/action/replication_record"}},"created_at":"2026-07-05T01:07:54.024627+00:00","updated_at":"2026-07-05T01:07:54.024627+00:00"}