{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:A7NOOQKB4MMA5XGTWB3Y5J5OJG","short_pith_number":"pith:A7NOOQKB","schema_version":"1.0","canonical_sha256":"07dae74141e3180edcd3b0778ea7ae4986df52d3c956c927a2fb9e0ca6144e10","source":{"kind":"arxiv","id":"2605.15355","version":1},"attestation_state":"computed","paper":{"title":"Federated Learning of Spiking Neural Networks under Heterogeneous Temporal Resolutions","license":"http://creativecommons.org/licenses/by/4.0/","headline":"Adaptation methods allow federated spiking neural networks to recover accuracy lost when clients sample data at different temporal resolutions.","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Ay\\c{c}a \\\"Oz\\c{c}elikkale, Sanja Karilanova, Subhrakanti Dey","submitted_at":"2026-05-14T19:33:35Z","abstract_excerpt":"Spiking neural networks (SNNs) are biologically inspired energy-efficient models that use sparse binary spike-based communication between neurons, making them attractive for resource-constrained edge devices. Federated learning enables such devices to train collaboratively without sharing raw data. In time-series applications, edge devices often collect data at different time resolutions due to hardware and energy constraints. This temporal heterogeneity poses a fundamental challenge for federated learning: parameters learned at one temporal resolution do not necessarily transfer directly to a"},"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":true,"formal_links_present":true},"canonical_record":{"source":{"id":"2605.15355","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-14T19:33:35Z","cross_cats_sorted":[],"title_canon_sha256":"07ccc9c749faaee69937c1b8d9c03ef469317b8fe7a30e89608d808041cf50fc","abstract_canon_sha256":"de4c51f743d3744384ffb6de587c92551c4bcf95d5e3a957460c6d5634bac908"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-20T00:00:54.121823Z","signature_b64":"QPKzSEcCSprWEwlAHVV0Xx6oX44Y6nYRsjuYJNZtkh80DdLOOD8T+Pd/L5CALO822P3meM8uV3HZ1Ds68vNcDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"07dae74141e3180edcd3b0778ea7ae4986df52d3c956c927a2fb9e0ca6144e10","last_reissued_at":"2026-05-20T00:00:54.121014Z","signature_status":"signed_v1","first_computed_at":"2026-05-20T00:00:54.121014Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Federated Learning of Spiking Neural Networks under Heterogeneous Temporal Resolutions","license":"http://creativecommons.org/licenses/by/4.0/","headline":"Adaptation methods allow federated spiking neural networks to recover accuracy lost when clients sample data at different temporal resolutions.","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Ay\\c{c}a \\\"Oz\\c{c}elikkale, Sanja Karilanova, Subhrakanti Dey","submitted_at":"2026-05-14T19:33:35Z","abstract_excerpt":"Spiking neural networks (SNNs) are biologically inspired energy-efficient models that use sparse binary spike-based communication between neurons, making them attractive for resource-constrained edge devices. Federated learning enables such devices to train collaboratively without sharing raw data. In time-series applications, edge devices often collect data at different time resolutions due to hardware and energy constraints. This temporal heterogeneity poses a fundamental challenge for federated learning: parameters learned at one temporal resolution do not necessarily transfer directly to a"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"The proposed adaptation methods can substantially recover accuracy lost due to temporal mismatch, hence enabling each client to train at their local temporal resolution while remaining compatible with the global model.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That neuron parameters learned at one temporal resolution can be meaningfully integrated with those from another resolution through the proposed aggregation rules without requiring changes to the underlying SNN architecture or loss of spike sparsity benefits.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"Federated learning framework for SNNs that adapts to heterogeneous temporal resolutions via neuron parameter integration, recovering accuracy on SHD and DVS-Gesture under varied mismatch scenarios.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Adaptation methods allow federated spiking neural networks to recover accuracy lost when clients sample data at different temporal resolutions.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"ff0bea6fc92f51152b9c592add41dce83bce675276a9b616b146c036aea93d09"},"source":{"id":"2605.15355","kind":"arxiv","version":1},"verdict":{"id":"acac3508-d693-4e49-8918-a4090e82a467","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-19T16:50:40.580981Z","strongest_claim":"The proposed adaptation methods can substantially recover accuracy lost due to temporal mismatch, hence enabling each client to train at their local temporal resolution while remaining compatible with the global model.","one_line_summary":"Federated learning framework for SNNs that adapts to heterogeneous temporal resolutions via neuron parameter integration, recovering accuracy on SHD and DVS-Gesture under varied mismatch scenarios.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That neuron parameters learned at one temporal resolution can be meaningfully integrated with those from another resolution through the proposed aggregation rules without requiring changes to the underlying SNN architecture or loss of spike sparsity benefits.","pith_extraction_headline":"Adaptation methods allow federated spiking neural networks to recover accuracy lost when clients sample data at different temporal resolutions."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2605.15355/integrity.json","findings":[],"available":true,"detectors_run":[{"name":"doi_title_agreement","ran_at":"2026-05-19T17:01:18.301664Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_compliance","ran_at":"2026-05-19T16:56:30.192927Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"claim_evidence","ran_at":"2026-05-19T14:21:54.199153Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"ai_meta_artifact","ran_at":"2026-05-19T13:33:22.746518Z","status":"skipped","version":"1.0.0","findings_count":0}],"snapshot_sha256":"69a86f5a6cc8d4e9f4bfe292e11b0c321371620b3125054e3a43284adccadbba"},"references":{"count":47,"sample":[{"doi":"","year":2021,"title":"Brendan McMahan","work_id":"363fd560-ca1d-4c04-a174-105f152b315b","ref_index":1,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2020,"title":"Parizi, and Fahad Saeed","work_id":"bdae9d7f-886c-48eb-ad67-edb4e61f0bc0","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2023,"title":"Data centers on wheels: Emissions from computing onboard autonomous vehicles.IEEE Micro, 43(1):29–39, 2023","work_id":"96f9b711-f5d9-43e1-b7f2-ef6452fdb3bf","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2019,"title":"Bipin Rajendran, Abu Sebastian, Michael Schmuker, Narayan Srinivasa, and Evangelos Eleft- heriou. Low-power neuromorphic hardware for signal processing applications: A review of architectural and syst","work_id":"eb44133c-19ea-4bc4-bde6-f85d83f9a3a4","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2021,"title":"Fonseca Guerra, Prasad Joshi, Philipp Plank, and Sumedh R","work_id":"764875c4-2cf9-48ba-9026-d9636efabe6a","ref_index":5,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":47,"snapshot_sha256":"65f6a07e63721ccc4f66493bf93c25daf35597e1d7b57734979c5110a820a183","internal_anchors":0},"formal_canon":{"evidence_count":2,"snapshot_sha256":"d3d1b27c6f86eb118818b6f0bf783ff44e6d1d5787c267a6678b898783ccc53c"},"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":"2605.15355","created_at":"2026-05-20T00:00:54.121140+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.15355v1","created_at":"2026-05-20T00:00:54.121140+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.15355","created_at":"2026-05-20T00:00:54.121140+00:00"},{"alias_kind":"pith_short_12","alias_value":"A7NOOQKB4MMA","created_at":"2026-05-20T00:00:54.121140+00:00"},{"alias_kind":"pith_short_16","alias_value":"A7NOOQKB4MMA5XGT","created_at":"2026-05-20T00:00:54.121140+00:00"},{"alias_kind":"pith_short_8","alias_value":"A7NOOQKB","created_at":"2026-05-20T00:00:54.121140+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":2,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/A7NOOQKB4MMA5XGTWB3Y5J5OJG","json":"https://pith.science/pith/A7NOOQKB4MMA5XGTWB3Y5J5OJG.json","graph_json":"https://pith.science/api/pith-number/A7NOOQKB4MMA5XGTWB3Y5J5OJG/graph.json","events_json":"https://pith.science/api/pith-number/A7NOOQKB4MMA5XGTWB3Y5J5OJG/events.json","paper":"https://pith.science/paper/A7NOOQKB"},"agent_actions":{"view_html":"https://pith.science/pith/A7NOOQKB4MMA5XGTWB3Y5J5OJG","download_json":"https://pith.science/pith/A7NOOQKB4MMA5XGTWB3Y5J5OJG.json","view_paper":"https://pith.science/paper/A7NOOQKB","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.15355&json=true","fetch_graph":"https://pith.science/api/pith-number/A7NOOQKB4MMA5XGTWB3Y5J5OJG/graph.json","fetch_events":"https://pith.science/api/pith-number/A7NOOQKB4MMA5XGTWB3Y5J5OJG/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/A7NOOQKB4MMA5XGTWB3Y5J5OJG/action/timestamp_anchor","attest_storage":"https://pith.science/pith/A7NOOQKB4MMA5XGTWB3Y5J5OJG/action/storage_attestation","attest_author":"https://pith.science/pith/A7NOOQKB4MMA5XGTWB3Y5J5OJG/action/author_attestation","sign_citation":"https://pith.science/pith/A7NOOQKB4MMA5XGTWB3Y5J5OJG/action/citation_signature","submit_replication":"https://pith.science/pith/A7NOOQKB4MMA5XGTWB3Y5J5OJG/action/replication_record"}},"created_at":"2026-05-20T00:00:54.121140+00:00","updated_at":"2026-05-20T00:00:54.121140+00:00"}