{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:PYYQ3SJHBNN2DCVRRVEI7GBYKF","short_pith_number":"pith:PYYQ3SJH","schema_version":"1.0","canonical_sha256":"7e310dc9270b5ba18ab18d488f98385143f0e201f108d1bb1a6e76c794eb8dbe","source":{"kind":"arxiv","id":"2106.08056","version":2},"attestation_state":"computed","paper":{"title":"Coupled Gradient Estimators for Discrete Latent Variables","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Andriy Mnih, George Tucker, Zhe Dong","submitted_at":"2021-06-15T11:28:44Z","abstract_excerpt":"Training models with discrete latent variables is challenging due to the high variance of unbiased gradient estimators. While low-variance reparameterization gradients of a continuous relaxation can provide an effective solution, a continuous relaxation is not always available or tractable. Dong et al. (2020) and Yin et al. (2020) introduced a performant estimator that does not rely on continuous relaxations; however, it is limited to binary random variables. We introduce a novel derivation of their estimator based on importance sampling and statistical couplings, which we extend to the catego"},"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":"2106.08056","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2021-06-15T11:28:44Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"1b2a50148187743b6f5b3124b717fad4e27bd3f49249fad23d1e2fb067033c5f","abstract_canon_sha256":"13d67fdc191cfb3af9cb923e44b96f011d92258bc72b75aafa26460cd7c70c2f"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:15:57.046981Z","signature_b64":"u5jT7x0QwjnaxK5leC4Su/N96Hsg4U8+N9EBfh3mKrzWkVdUTDsSlvkFWULB91Oz+9N/a3P606pFF8FkuCDkDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7e310dc9270b5ba18ab18d488f98385143f0e201f108d1bb1a6e76c794eb8dbe","last_reissued_at":"2026-07-05T05:15:57.046624Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:15:57.046624Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Coupled Gradient Estimators for Discrete Latent Variables","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Andriy Mnih, George Tucker, Zhe Dong","submitted_at":"2021-06-15T11:28:44Z","abstract_excerpt":"Training models with discrete latent variables is challenging due to the high variance of unbiased gradient estimators. While low-variance reparameterization gradients of a continuous relaxation can provide an effective solution, a continuous relaxation is not always available or tractable. Dong et al. (2020) and Yin et al. (2020) introduced a performant estimator that does not rely on continuous relaxations; however, it is limited to binary random variables. We introduce a novel derivation of their estimator based on importance sampling and statistical couplings, which we extend to the catego"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2106.08056","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/2106.08056/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":"2106.08056","created_at":"2026-07-05T05:15:57.046689+00:00"},{"alias_kind":"arxiv_version","alias_value":"2106.08056v2","created_at":"2026-07-05T05:15:57.046689+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2106.08056","created_at":"2026-07-05T05:15:57.046689+00:00"},{"alias_kind":"pith_short_12","alias_value":"PYYQ3SJHBNN2","created_at":"2026-07-05T05:15:57.046689+00:00"},{"alias_kind":"pith_short_16","alias_value":"PYYQ3SJHBNN2DCVR","created_at":"2026-07-05T05:15:57.046689+00:00"},{"alias_kind":"pith_short_8","alias_value":"PYYQ3SJH","created_at":"2026-07-05T05:15:57.046689+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/PYYQ3SJHBNN2DCVRRVEI7GBYKF","json":"https://pith.science/pith/PYYQ3SJHBNN2DCVRRVEI7GBYKF.json","graph_json":"https://pith.science/api/pith-number/PYYQ3SJHBNN2DCVRRVEI7GBYKF/graph.json","events_json":"https://pith.science/api/pith-number/PYYQ3SJHBNN2DCVRRVEI7GBYKF/events.json","paper":"https://pith.science/paper/PYYQ3SJH"},"agent_actions":{"view_html":"https://pith.science/pith/PYYQ3SJHBNN2DCVRRVEI7GBYKF","download_json":"https://pith.science/pith/PYYQ3SJHBNN2DCVRRVEI7GBYKF.json","view_paper":"https://pith.science/paper/PYYQ3SJH","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2106.08056&json=true","fetch_graph":"https://pith.science/api/pith-number/PYYQ3SJHBNN2DCVRRVEI7GBYKF/graph.json","fetch_events":"https://pith.science/api/pith-number/PYYQ3SJHBNN2DCVRRVEI7GBYKF/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/PYYQ3SJHBNN2DCVRRVEI7GBYKF/action/timestamp_anchor","attest_storage":"https://pith.science/pith/PYYQ3SJHBNN2DCVRRVEI7GBYKF/action/storage_attestation","attest_author":"https://pith.science/pith/PYYQ3SJHBNN2DCVRRVEI7GBYKF/action/author_attestation","sign_citation":"https://pith.science/pith/PYYQ3SJHBNN2DCVRRVEI7GBYKF/action/citation_signature","submit_replication":"https://pith.science/pith/PYYQ3SJHBNN2DCVRRVEI7GBYKF/action/replication_record"}},"created_at":"2026-07-05T05:15:57.046689+00:00","updated_at":"2026-07-05T05:15:57.046689+00:00"}