{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:MLJNS4RZXE7I6S25LBIP6ZSZXE","short_pith_number":"pith:MLJNS4RZ","schema_version":"1.0","canonical_sha256":"62d2d97239b93e8f4b5d5850ff6659b912c2961eaf6b40b7bd7958a2cc79b189","source":{"kind":"arxiv","id":"2501.13181","version":1},"attestation_state":"computed","paper":{"title":"Learning in Log-Domain: Subthreshold Analog AI Accelerator Based on Stochastic Gradient Descent","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.AR","authors_text":"Emmanuel M Drakakis, Momen K Tageldeen, Vivek Mohan, Yacine Belgaid, Zhou Wang","submitted_at":"2025-01-22T19:26:36Z","abstract_excerpt":"The rapid proliferation of AI models, coupled with growing demand for edge deployment, necessitates the development of AI hardware that is both high-performance and energy-efficient. In this paper, we propose a novel analog accelerator architecture designed for AI/ML training workloads using stochastic gradient descent with L2 regularization (SGDr). The architecture leverages log-domain circuits in subthreshold MOS and incorporates volatile memory. We establish a mathematical framework for solving SGDr in the continuous time domain and detail the mapping of SGDr learning equations to log-domai"},"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":"2501.13181","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AR","submitted_at":"2025-01-22T19:26:36Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"2edcbe2f35604b2128187b787f2599e9b86b1f93f3a1ace6edf56c64e48123e3","abstract_canon_sha256":"5fff630face382749ff7fa86b85c10fb720c2f726794f7bd1832891fd8802d09"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:04:24.694823Z","signature_b64":"34axSGPEURICu2b1Vq1ymo5itUwagClWKhGbsmSQZH4TCvM8Y5kaHztmiZysHDMO2bSVIc35uDZKAk41dfinCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"62d2d97239b93e8f4b5d5850ff6659b912c2961eaf6b40b7bd7958a2cc79b189","last_reissued_at":"2026-07-05T10:04:24.694279Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:04:24.694279Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Learning in Log-Domain: Subthreshold Analog AI Accelerator Based on Stochastic Gradient Descent","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.AR","authors_text":"Emmanuel M Drakakis, Momen K Tageldeen, Vivek Mohan, Yacine Belgaid, Zhou Wang","submitted_at":"2025-01-22T19:26:36Z","abstract_excerpt":"The rapid proliferation of AI models, coupled with growing demand for edge deployment, necessitates the development of AI hardware that is both high-performance and energy-efficient. In this paper, we propose a novel analog accelerator architecture designed for AI/ML training workloads using stochastic gradient descent with L2 regularization (SGDr). The architecture leverages log-domain circuits in subthreshold MOS and incorporates volatile memory. We establish a mathematical framework for solving SGDr in the continuous time domain and detail the mapping of SGDr learning equations to log-domai"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2501.13181","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/2501.13181/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":"2501.13181","created_at":"2026-07-05T10:04:24.694346+00:00"},{"alias_kind":"arxiv_version","alias_value":"2501.13181v1","created_at":"2026-07-05T10:04:24.694346+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2501.13181","created_at":"2026-07-05T10:04:24.694346+00:00"},{"alias_kind":"pith_short_12","alias_value":"MLJNS4RZXE7I","created_at":"2026-07-05T10:04:24.694346+00:00"},{"alias_kind":"pith_short_16","alias_value":"MLJNS4RZXE7I6S25","created_at":"2026-07-05T10:04:24.694346+00:00"},{"alias_kind":"pith_short_8","alias_value":"MLJNS4RZ","created_at":"2026-07-05T10:04:24.694346+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/MLJNS4RZXE7I6S25LBIP6ZSZXE","json":"https://pith.science/pith/MLJNS4RZXE7I6S25LBIP6ZSZXE.json","graph_json":"https://pith.science/api/pith-number/MLJNS4RZXE7I6S25LBIP6ZSZXE/graph.json","events_json":"https://pith.science/api/pith-number/MLJNS4RZXE7I6S25LBIP6ZSZXE/events.json","paper":"https://pith.science/paper/MLJNS4RZ"},"agent_actions":{"view_html":"https://pith.science/pith/MLJNS4RZXE7I6S25LBIP6ZSZXE","download_json":"https://pith.science/pith/MLJNS4RZXE7I6S25LBIP6ZSZXE.json","view_paper":"https://pith.science/paper/MLJNS4RZ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2501.13181&json=true","fetch_graph":"https://pith.science/api/pith-number/MLJNS4RZXE7I6S25LBIP6ZSZXE/graph.json","fetch_events":"https://pith.science/api/pith-number/MLJNS4RZXE7I6S25LBIP6ZSZXE/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/MLJNS4RZXE7I6S25LBIP6ZSZXE/action/timestamp_anchor","attest_storage":"https://pith.science/pith/MLJNS4RZXE7I6S25LBIP6ZSZXE/action/storage_attestation","attest_author":"https://pith.science/pith/MLJNS4RZXE7I6S25LBIP6ZSZXE/action/author_attestation","sign_citation":"https://pith.science/pith/MLJNS4RZXE7I6S25LBIP6ZSZXE/action/citation_signature","submit_replication":"https://pith.science/pith/MLJNS4RZXE7I6S25LBIP6ZSZXE/action/replication_record"}},"created_at":"2026-07-05T10:04:24.694346+00:00","updated_at":"2026-07-05T10:04:24.694346+00:00"}