{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:TMBYFKEGPCQVHTNHL5GLKZSVFF","short_pith_number":"pith:TMBYFKEG","schema_version":"1.0","canonical_sha256":"9b0382a88678a153cda75f4cb56655294c04f3f31e7e7c6589fcb228b7cba6e0","source":{"kind":"arxiv","id":"2409.18566","version":2},"attestation_state":"computed","paper":{"title":"Optimizing DNN Inference on Multi-Accelerator SoCs at Training-time","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Alessio Burrello, Daniele Jahier Pagliari, Matteo Risso","submitted_at":"2024-09-27T09:10:44Z","abstract_excerpt":"The demand for executing Deep Neural Networks (DNNs) with low latency and minimal power consumption at the edge has led to the development of advanced heterogeneous Systems-on-Chips (SoCs) that incorporate multiple specialized computing units (CUs), such as accelerators. Offloading DNN computations to a specific CU from the available set often exposes accuracy vs efficiency trade-offs, due to differences in their supported operations (e.g., standard vs. depthwise convolution) or data representations (e.g., more/less aggressively quantized). A challenging yet unresolved issue is how to map a DN"},"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":"2409.18566","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-09-27T09:10:44Z","cross_cats_sorted":[],"title_canon_sha256":"82843d49409d4fb52697817605ca8cd7010f7ae2c0e251f8b132c18184e836bd","abstract_canon_sha256":"cf054922f95b5edaf8da0ccb5f0f7bb3b73bbcb529592962e6d941be1428f22a"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:17:49.905514Z","signature_b64":"Ac3cs4AoxRkZRCisCr0SJ3VTJwJI5WKhaeZvQ2j8/ByO8u50GGH396187WqPg8hVfTof4QFG55XClRD7lIlXBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9b0382a88678a153cda75f4cb56655294c04f3f31e7e7c6589fcb228b7cba6e0","last_reissued_at":"2026-07-05T10:17:49.905018Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:17:49.905018Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Optimizing DNN Inference on Multi-Accelerator SoCs at Training-time","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Alessio Burrello, Daniele Jahier Pagliari, Matteo Risso","submitted_at":"2024-09-27T09:10:44Z","abstract_excerpt":"The demand for executing Deep Neural Networks (DNNs) with low latency and minimal power consumption at the edge has led to the development of advanced heterogeneous Systems-on-Chips (SoCs) that incorporate multiple specialized computing units (CUs), such as accelerators. Offloading DNN computations to a specific CU from the available set often exposes accuracy vs efficiency trade-offs, due to differences in their supported operations (e.g., standard vs. depthwise convolution) or data representations (e.g., more/less aggressively quantized). A challenging yet unresolved issue is how to map a DN"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2409.18566","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/2409.18566/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":"2409.18566","created_at":"2026-07-05T10:17:49.905092+00:00"},{"alias_kind":"arxiv_version","alias_value":"2409.18566v2","created_at":"2026-07-05T10:17:49.905092+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2409.18566","created_at":"2026-07-05T10:17:49.905092+00:00"},{"alias_kind":"pith_short_12","alias_value":"TMBYFKEGPCQV","created_at":"2026-07-05T10:17:49.905092+00:00"},{"alias_kind":"pith_short_16","alias_value":"TMBYFKEGPCQVHTNH","created_at":"2026-07-05T10:17:49.905092+00:00"},{"alias_kind":"pith_short_8","alias_value":"TMBYFKEG","created_at":"2026-07-05T10:17:49.905092+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/TMBYFKEGPCQVHTNHL5GLKZSVFF","json":"https://pith.science/pith/TMBYFKEGPCQVHTNHL5GLKZSVFF.json","graph_json":"https://pith.science/api/pith-number/TMBYFKEGPCQVHTNHL5GLKZSVFF/graph.json","events_json":"https://pith.science/api/pith-number/TMBYFKEGPCQVHTNHL5GLKZSVFF/events.json","paper":"https://pith.science/paper/TMBYFKEG"},"agent_actions":{"view_html":"https://pith.science/pith/TMBYFKEGPCQVHTNHL5GLKZSVFF","download_json":"https://pith.science/pith/TMBYFKEGPCQVHTNHL5GLKZSVFF.json","view_paper":"https://pith.science/paper/TMBYFKEG","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2409.18566&json=true","fetch_graph":"https://pith.science/api/pith-number/TMBYFKEGPCQVHTNHL5GLKZSVFF/graph.json","fetch_events":"https://pith.science/api/pith-number/TMBYFKEGPCQVHTNHL5GLKZSVFF/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/TMBYFKEGPCQVHTNHL5GLKZSVFF/action/timestamp_anchor","attest_storage":"https://pith.science/pith/TMBYFKEGPCQVHTNHL5GLKZSVFF/action/storage_attestation","attest_author":"https://pith.science/pith/TMBYFKEGPCQVHTNHL5GLKZSVFF/action/author_attestation","sign_citation":"https://pith.science/pith/TMBYFKEGPCQVHTNHL5GLKZSVFF/action/citation_signature","submit_replication":"https://pith.science/pith/TMBYFKEGPCQVHTNHL5GLKZSVFF/action/replication_record"}},"created_at":"2026-07-05T10:17:49.905092+00:00","updated_at":"2026-07-05T10:17:49.905092+00:00"}