{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:SBRZRFIL65RPPIY5PT7WYQ2KGE","short_pith_number":"pith:SBRZRFIL","schema_version":"1.0","canonical_sha256":"906398950bf762f7a31d7cff6c434a312576ee89d02191a8131ae1d421805772","source":{"kind":"arxiv","id":"2305.14974","version":2},"attestation_state":"computed","paper":{"title":"Block-local learning with probabilistic latent representations","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Anand Subramoney, Cabrel Teguemne Fokam, Christian Mayr, David Kappel, Khaleelulla Khan Nazeer","submitted_at":"2023-05-24T10:11:30Z","abstract_excerpt":"The ubiquitous backpropagation algorithm requires sequential updates through the network introducing a locking problem. In addition, back-propagation relies on the transpose of forward weight matrices to compute updates, introducing a weight transport problem across the network. Locking and weight transport are problems because they prevent efficient parallelization and horizontal scaling of the training process. We propose a new method to address both these problems and scale up the training of large models. Our method works by dividing a deep neural network into blocks and introduces a feedb"},"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":"2305.14974","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2023-05-24T10:11:30Z","cross_cats_sorted":[],"title_canon_sha256":"5e896c1ea73a4840bb6b295e15b98c966700a0337c2e3471796403fd921ae099","abstract_canon_sha256":"aaeea645606374eaf16c4a375d1c0598f26f6c315d45078ca0439fdb8dcb36cb"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:06:03.250643Z","signature_b64":"m/5cNLADupybmQzWPvpFLweiPs57c399yQKM/JbSqNnf0DE6+QDT1jqd6Uq0UTlt4wPGBcscVSvRrb9RwPTxDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"906398950bf762f7a31d7cff6c434a312576ee89d02191a8131ae1d421805772","last_reissued_at":"2026-07-05T07:06:03.250182Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:06:03.250182Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Block-local learning with probabilistic latent representations","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Anand Subramoney, Cabrel Teguemne Fokam, Christian Mayr, David Kappel, Khaleelulla Khan Nazeer","submitted_at":"2023-05-24T10:11:30Z","abstract_excerpt":"The ubiquitous backpropagation algorithm requires sequential updates through the network introducing a locking problem. In addition, back-propagation relies on the transpose of forward weight matrices to compute updates, introducing a weight transport problem across the network. Locking and weight transport are problems because they prevent efficient parallelization and horizontal scaling of the training process. We propose a new method to address both these problems and scale up the training of large models. Our method works by dividing a deep neural network into blocks and introduces a feedb"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2305.14974","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/2305.14974/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":"2305.14974","created_at":"2026-07-05T07:06:03.250239+00:00"},{"alias_kind":"arxiv_version","alias_value":"2305.14974v2","created_at":"2026-07-05T07:06:03.250239+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2305.14974","created_at":"2026-07-05T07:06:03.250239+00:00"},{"alias_kind":"pith_short_12","alias_value":"SBRZRFIL65RP","created_at":"2026-07-05T07:06:03.250239+00:00"},{"alias_kind":"pith_short_16","alias_value":"SBRZRFIL65RPPIY5","created_at":"2026-07-05T07:06:03.250239+00:00"},{"alias_kind":"pith_short_8","alias_value":"SBRZRFIL","created_at":"2026-07-05T07:06:03.250239+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.21497","citing_title":"Breaking chains with trees: Deep learning with $\\mathcal{O}(\\log N)$ parallel time complexity","ref_index":6,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/SBRZRFIL65RPPIY5PT7WYQ2KGE","json":"https://pith.science/pith/SBRZRFIL65RPPIY5PT7WYQ2KGE.json","graph_json":"https://pith.science/api/pith-number/SBRZRFIL65RPPIY5PT7WYQ2KGE/graph.json","events_json":"https://pith.science/api/pith-number/SBRZRFIL65RPPIY5PT7WYQ2KGE/events.json","paper":"https://pith.science/paper/SBRZRFIL"},"agent_actions":{"view_html":"https://pith.science/pith/SBRZRFIL65RPPIY5PT7WYQ2KGE","download_json":"https://pith.science/pith/SBRZRFIL65RPPIY5PT7WYQ2KGE.json","view_paper":"https://pith.science/paper/SBRZRFIL","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2305.14974&json=true","fetch_graph":"https://pith.science/api/pith-number/SBRZRFIL65RPPIY5PT7WYQ2KGE/graph.json","fetch_events":"https://pith.science/api/pith-number/SBRZRFIL65RPPIY5PT7WYQ2KGE/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/SBRZRFIL65RPPIY5PT7WYQ2KGE/action/timestamp_anchor","attest_storage":"https://pith.science/pith/SBRZRFIL65RPPIY5PT7WYQ2KGE/action/storage_attestation","attest_author":"https://pith.science/pith/SBRZRFIL65RPPIY5PT7WYQ2KGE/action/author_attestation","sign_citation":"https://pith.science/pith/SBRZRFIL65RPPIY5PT7WYQ2KGE/action/citation_signature","submit_replication":"https://pith.science/pith/SBRZRFIL65RPPIY5PT7WYQ2KGE/action/replication_record"}},"created_at":"2026-07-05T07:06:03.250239+00:00","updated_at":"2026-07-05T07:06:03.250239+00:00"}