{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:VLLHXNEWQAPE2NL46U5TQR6OMI","short_pith_number":"pith:VLLHXNEW","schema_version":"1.0","canonical_sha256":"aad67bb496801e4d357cf53b3847ce6218dae912ec06e235509f5ae929b1d7de","source":{"kind":"arxiv","id":"2105.00395","version":1},"attestation_state":"computed","paper":{"title":"AirMixML: Over-the-Air Data Mixup for Inherently Privacy-Preserving Edge Machine Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CR","cs.LG"],"primary_cat":"cs.NI","authors_text":"Jihong Park, Mehdi Bennis, Praneeth Vepakomma, Ramesh Raskar, Yusuke Koda","submitted_at":"2021-05-02T05:45:43Z","abstract_excerpt":"Wireless channels can be inherently privacy-preserving by distorting the received signals due to channel noise, and superpositioning multiple signals over-the-air. By harnessing these natural distortions and superpositions by wireless channels, we propose a novel privacy-preserving machine learning (ML) framework at the network edge, coined over-the-air mixup ML (AirMixML). In AirMixML, multiple workers transmit analog-modulated signals of their private data samples to an edge server who trains an ML model using the received noisy-and superpositioned samples. AirMixML coincides with model trai"},"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":"2105.00395","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.NI","submitted_at":"2021-05-02T05:45:43Z","cross_cats_sorted":["cs.AI","cs.CR","cs.LG"],"title_canon_sha256":"e69d703cf31a752722d3594df2b9d020da2687cd6f8eaf048a7b02ff99ac87b4","abstract_canon_sha256":"af89d4fb49f38183771ad12f877175010faedfc91ee9870661bb8f653fa1163c"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:36:54.069231Z","signature_b64":"kwYZM7jk6DyGeDXLcCX9fb1fhiXNaVzWHc5EKc1uo/D8wpeosPBDB416MT882XVmakYT0V0lW/A4xOrnVUHIAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"aad67bb496801e4d357cf53b3847ce6218dae912ec06e235509f5ae929b1d7de","last_reissued_at":"2026-07-05T02:36:54.068792Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:36:54.068792Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"AirMixML: Over-the-Air Data Mixup for Inherently Privacy-Preserving Edge Machine Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CR","cs.LG"],"primary_cat":"cs.NI","authors_text":"Jihong Park, Mehdi Bennis, Praneeth Vepakomma, Ramesh Raskar, Yusuke Koda","submitted_at":"2021-05-02T05:45:43Z","abstract_excerpt":"Wireless channels can be inherently privacy-preserving by distorting the received signals due to channel noise, and superpositioning multiple signals over-the-air. By harnessing these natural distortions and superpositions by wireless channels, we propose a novel privacy-preserving machine learning (ML) framework at the network edge, coined over-the-air mixup ML (AirMixML). In AirMixML, multiple workers transmit analog-modulated signals of their private data samples to an edge server who trains an ML model using the received noisy-and superpositioned samples. AirMixML coincides with model trai"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2105.00395","kind":"arxiv","version":1},"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/2105.00395/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":"2105.00395","created_at":"2026-07-05T02:36:54.068852+00:00"},{"alias_kind":"arxiv_version","alias_value":"2105.00395v1","created_at":"2026-07-05T02:36:54.068852+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2105.00395","created_at":"2026-07-05T02:36:54.068852+00:00"},{"alias_kind":"pith_short_12","alias_value":"VLLHXNEWQAPE","created_at":"2026-07-05T02:36:54.068852+00:00"},{"alias_kind":"pith_short_16","alias_value":"VLLHXNEWQAPE2NL4","created_at":"2026-07-05T02:36:54.068852+00:00"},{"alias_kind":"pith_short_8","alias_value":"VLLHXNEW","created_at":"2026-07-05T02:36:54.068852+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2411.15948","citing_title":"Over-the-Air Federated Adaptive Data Analysis: Preserving Accuracy via Opportunistic Differential Privacy","ref_index":12,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/VLLHXNEWQAPE2NL46U5TQR6OMI","json":"https://pith.science/pith/VLLHXNEWQAPE2NL46U5TQR6OMI.json","graph_json":"https://pith.science/api/pith-number/VLLHXNEWQAPE2NL46U5TQR6OMI/graph.json","events_json":"https://pith.science/api/pith-number/VLLHXNEWQAPE2NL46U5TQR6OMI/events.json","paper":"https://pith.science/paper/VLLHXNEW"},"agent_actions":{"view_html":"https://pith.science/pith/VLLHXNEWQAPE2NL46U5TQR6OMI","download_json":"https://pith.science/pith/VLLHXNEWQAPE2NL46U5TQR6OMI.json","view_paper":"https://pith.science/paper/VLLHXNEW","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2105.00395&json=true","fetch_graph":"https://pith.science/api/pith-number/VLLHXNEWQAPE2NL46U5TQR6OMI/graph.json","fetch_events":"https://pith.science/api/pith-number/VLLHXNEWQAPE2NL46U5TQR6OMI/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/VLLHXNEWQAPE2NL46U5TQR6OMI/action/timestamp_anchor","attest_storage":"https://pith.science/pith/VLLHXNEWQAPE2NL46U5TQR6OMI/action/storage_attestation","attest_author":"https://pith.science/pith/VLLHXNEWQAPE2NL46U5TQR6OMI/action/author_attestation","sign_citation":"https://pith.science/pith/VLLHXNEWQAPE2NL46U5TQR6OMI/action/citation_signature","submit_replication":"https://pith.science/pith/VLLHXNEWQAPE2NL46U5TQR6OMI/action/replication_record"}},"created_at":"2026-07-05T02:36:54.068852+00:00","updated_at":"2026-07-05T02:36:54.068852+00:00"}