{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2018:4SZUETUBMBRQ5HC4GC77EQMOB4","short_pith_number":"pith:4SZUETUB","schema_version":"1.0","canonical_sha256":"e4b3424e8160630e9c5c30bff2418e0f0776e2fca7e3e081055f0a6ebe0045fd","source":{"kind":"arxiv","id":"1807.01406","version":2},"attestation_state":"computed","paper":{"title":"Connecting Weighted Automata and Recurrent Neural Networks through Spectral Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.FL","stat.ML"],"primary_cat":"cs.LG","authors_text":"Doina Precup, Guillaume Rabusseau, Tianyu Li","submitted_at":"2018-07-04T00:12:12Z","abstract_excerpt":"In this paper, we unravel a fundamental connection between weighted finite automata~(WFAs) and second-order recurrent neural networks~(2-RNNs): in the case of sequences of discrete symbols, WFAs and 2-RNNs with linear activation functions are expressively equivalent. Motivated by this result, we build upon a recent extension of the spectral learning algorithm to vector-valued WFAs and propose the first provable learning algorithm for linear 2-RNNs defined over sequences of continuous input vectors. This algorithm relies on estimating low rank sub-blocks of the so-called Hankel tensor, from whi"},"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":"1807.01406","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-07-04T00:12:12Z","cross_cats_sorted":["cs.FL","stat.ML"],"title_canon_sha256":"76bc7f94836030ab94a0482903be6ef467307ac65b53cf64698e36a938248e31","abstract_canon_sha256":"4db479733252212b41f05912fc8864f6699fe2234928b7f3001da2104b332607"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:49:11.581925Z","signature_b64":"YWsoo2Y7rnGvzbrA4Xz5IRKKkJZkVGXCYto4JBAR8tYts0ZWuUsQSx5eWxR3d6iNdE5RhscOTSfL4pgfmDxlBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e4b3424e8160630e9c5c30bff2418e0f0776e2fca7e3e081055f0a6ebe0045fd","last_reissued_at":"2026-05-17T23:49:11.581236Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:49:11.581236Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Connecting Weighted Automata and Recurrent Neural Networks through Spectral Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.FL","stat.ML"],"primary_cat":"cs.LG","authors_text":"Doina Precup, Guillaume Rabusseau, Tianyu Li","submitted_at":"2018-07-04T00:12:12Z","abstract_excerpt":"In this paper, we unravel a fundamental connection between weighted finite automata~(WFAs) and second-order recurrent neural networks~(2-RNNs): in the case of sequences of discrete symbols, WFAs and 2-RNNs with linear activation functions are expressively equivalent. Motivated by this result, we build upon a recent extension of the spectral learning algorithm to vector-valued WFAs and propose the first provable learning algorithm for linear 2-RNNs defined over sequences of continuous input vectors. This algorithm relies on estimating low rank sub-blocks of the so-called Hankel tensor, from whi"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1807.01406","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":""},"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":"1807.01406","created_at":"2026-05-17T23:49:11.581342+00:00"},{"alias_kind":"arxiv_version","alias_value":"1807.01406v2","created_at":"2026-05-17T23:49:11.581342+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1807.01406","created_at":"2026-05-17T23:49:11.581342+00:00"},{"alias_kind":"pith_short_12","alias_value":"4SZUETUBMBRQ","created_at":"2026-05-18T12:32:05.422762+00:00"},{"alias_kind":"pith_short_16","alias_value":"4SZUETUBMBRQ5HC4","created_at":"2026-05-18T12:32:05.422762+00:00"},{"alias_kind":"pith_short_8","alias_value":"4SZUETUB","created_at":"2026-05-18T12:32:05.422762+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/4SZUETUBMBRQ5HC4GC77EQMOB4","json":"https://pith.science/pith/4SZUETUBMBRQ5HC4GC77EQMOB4.json","graph_json":"https://pith.science/api/pith-number/4SZUETUBMBRQ5HC4GC77EQMOB4/graph.json","events_json":"https://pith.science/api/pith-number/4SZUETUBMBRQ5HC4GC77EQMOB4/events.json","paper":"https://pith.science/paper/4SZUETUB"},"agent_actions":{"view_html":"https://pith.science/pith/4SZUETUBMBRQ5HC4GC77EQMOB4","download_json":"https://pith.science/pith/4SZUETUBMBRQ5HC4GC77EQMOB4.json","view_paper":"https://pith.science/paper/4SZUETUB","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1807.01406&json=true","fetch_graph":"https://pith.science/api/pith-number/4SZUETUBMBRQ5HC4GC77EQMOB4/graph.json","fetch_events":"https://pith.science/api/pith-number/4SZUETUBMBRQ5HC4GC77EQMOB4/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/4SZUETUBMBRQ5HC4GC77EQMOB4/action/timestamp_anchor","attest_storage":"https://pith.science/pith/4SZUETUBMBRQ5HC4GC77EQMOB4/action/storage_attestation","attest_author":"https://pith.science/pith/4SZUETUBMBRQ5HC4GC77EQMOB4/action/author_attestation","sign_citation":"https://pith.science/pith/4SZUETUBMBRQ5HC4GC77EQMOB4/action/citation_signature","submit_replication":"https://pith.science/pith/4SZUETUBMBRQ5HC4GC77EQMOB4/action/replication_record"}},"created_at":"2026-05-17T23:49:11.581342+00:00","updated_at":"2026-05-17T23:49:11.581342+00:00"}