{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:WFTTVI2P4R7XXLMBJ7MZPCWMZJ","short_pith_number":"pith:WFTTVI2P","schema_version":"1.0","canonical_sha256":"b1673aa34fe47f7bad814fd9978accca70ec4a7a9015785378de63ca83340049","source":{"kind":"arxiv","id":"2203.03324","version":1},"attestation_state":"computed","paper":{"title":"Dynamic ConvNets on Tiny Devices via Nested Sparsity","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Andrea Calimera, Antonio Cipolletta, Luca Mocerino, Matteo Grimaldi","submitted_at":"2022-03-07T12:07:02Z","abstract_excerpt":"This work introduces a new training and compression pipeline to build Nested Sparse ConvNets, a class of dynamic Convolutional Neural Networks (ConvNets) suited for inference tasks deployed on resource-constrained devices at the edge of the Internet-of-Things. A Nested Sparse ConvNet consists of a single ConvNet architecture containing N sparse sub-networks with nested weights subsets, like a Matryoshka doll, and can trade accuracy for latency at run time, using the model sparsity as a dynamic knob. To attain high accuracy at training time, we propose a gradient masking technique that optimall"},"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":"2203.03324","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2022-03-07T12:07:02Z","cross_cats_sorted":[],"title_canon_sha256":"59091fd5bfdc752252af52f78b04939ad8df23719800e61a9cbee5055eba521b","abstract_canon_sha256":"e9b3346e5f2cf39abca27c16abe6c66720c77d9537e7fd54043e7fb733ac3c64"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:02:33.433958Z","signature_b64":"O+0Rn73PjeIV4mSgfXqARmlajTxvckstsCUxtwojDH39YexLaCW/Jp1pX05kpVFstwc8xedxLbBrdNOwhMwkAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b1673aa34fe47f7bad814fd9978accca70ec4a7a9015785378de63ca83340049","last_reissued_at":"2026-07-05T04:02:33.433565Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:02:33.433565Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Dynamic ConvNets on Tiny Devices via Nested Sparsity","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Andrea Calimera, Antonio Cipolletta, Luca Mocerino, Matteo Grimaldi","submitted_at":"2022-03-07T12:07:02Z","abstract_excerpt":"This work introduces a new training and compression pipeline to build Nested Sparse ConvNets, a class of dynamic Convolutional Neural Networks (ConvNets) suited for inference tasks deployed on resource-constrained devices at the edge of the Internet-of-Things. A Nested Sparse ConvNet consists of a single ConvNet architecture containing N sparse sub-networks with nested weights subsets, like a Matryoshka doll, and can trade accuracy for latency at run time, using the model sparsity as a dynamic knob. To attain high accuracy at training time, we propose a gradient masking technique that optimall"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2203.03324","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/2203.03324/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":"2203.03324","created_at":"2026-07-05T04:02:33.433628+00:00"},{"alias_kind":"arxiv_version","alias_value":"2203.03324v1","created_at":"2026-07-05T04:02:33.433628+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2203.03324","created_at":"2026-07-05T04:02:33.433628+00:00"},{"alias_kind":"pith_short_12","alias_value":"WFTTVI2P4R7X","created_at":"2026-07-05T04:02:33.433628+00:00"},{"alias_kind":"pith_short_16","alias_value":"WFTTVI2P4R7XXLMB","created_at":"2026-07-05T04:02:33.433628+00:00"},{"alias_kind":"pith_short_8","alias_value":"WFTTVI2P","created_at":"2026-07-05T04:02:33.433628+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/WFTTVI2P4R7XXLMBJ7MZPCWMZJ","json":"https://pith.science/pith/WFTTVI2P4R7XXLMBJ7MZPCWMZJ.json","graph_json":"https://pith.science/api/pith-number/WFTTVI2P4R7XXLMBJ7MZPCWMZJ/graph.json","events_json":"https://pith.science/api/pith-number/WFTTVI2P4R7XXLMBJ7MZPCWMZJ/events.json","paper":"https://pith.science/paper/WFTTVI2P"},"agent_actions":{"view_html":"https://pith.science/pith/WFTTVI2P4R7XXLMBJ7MZPCWMZJ","download_json":"https://pith.science/pith/WFTTVI2P4R7XXLMBJ7MZPCWMZJ.json","view_paper":"https://pith.science/paper/WFTTVI2P","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2203.03324&json=true","fetch_graph":"https://pith.science/api/pith-number/WFTTVI2P4R7XXLMBJ7MZPCWMZJ/graph.json","fetch_events":"https://pith.science/api/pith-number/WFTTVI2P4R7XXLMBJ7MZPCWMZJ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/WFTTVI2P4R7XXLMBJ7MZPCWMZJ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/WFTTVI2P4R7XXLMBJ7MZPCWMZJ/action/storage_attestation","attest_author":"https://pith.science/pith/WFTTVI2P4R7XXLMBJ7MZPCWMZJ/action/author_attestation","sign_citation":"https://pith.science/pith/WFTTVI2P4R7XXLMBJ7MZPCWMZJ/action/citation_signature","submit_replication":"https://pith.science/pith/WFTTVI2P4R7XXLMBJ7MZPCWMZJ/action/replication_record"}},"created_at":"2026-07-05T04:02:33.433628+00:00","updated_at":"2026-07-05T04:02:33.433628+00:00"}