{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2019:AO3EY7PSAQ6L55OGX3GZK4KOFS","short_pith_number":"pith:AO3EY7PS","schema_version":"1.0","canonical_sha256":"03b64c7df2043cbef5c6becd95714e2caec5e65bfb4b2cf1d27cebf07bd1ab67","source":{"kind":"arxiv","id":"1911.10036","version":1},"attestation_state":"computed","paper":{"title":"Low-variance Black-box Gradient Estimates for the Plackett-Luce Distribution","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Artyom Gadetsky, Christopher Robinson, Dmitry Vetrov, Kirill Struminsky, Novi Quadrianto","submitted_at":"2019-11-22T13:34:50Z","abstract_excerpt":"Learning models with discrete latent variables using stochastic gradient descent remains a challenge due to the high variance of gradient estimates. Modern variance reduction techniques mostly consider categorical distributions and have limited applicability when the number of possible outcomes becomes large. In this work, we consider models with latent permutations and propose control variates for the Plackett-Luce distribution. In particular, the control variates allow us to optimize black-box functions over permutations using stochastic gradient descent. To illustrate the approach, we consi"},"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":"1911.10036","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-11-22T13:34:50Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"f3a8b5a9a30c05294a657f2139f1bf0fae10c9a606c38df656968f8cf3c2d6ca","abstract_canon_sha256":"1f91564ba3ee29f59c06f567f8490abdf465c251e91c3b21903999f70425cc22"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T00:21:19.939714Z","signature_b64":"vsiHnREZzsv8F8RSWO+T8DNaHZ3bgx2PRbSp+B+oInBj7+FnftF6gLp3R9VZ/IP4RkLDEt6TFa9ODrJINXfUBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"03b64c7df2043cbef5c6becd95714e2caec5e65bfb4b2cf1d27cebf07bd1ab67","last_reissued_at":"2026-07-05T00:21:19.939375Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T00:21:19.939375Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Low-variance Black-box Gradient Estimates for the Plackett-Luce Distribution","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Artyom Gadetsky, Christopher Robinson, Dmitry Vetrov, Kirill Struminsky, Novi Quadrianto","submitted_at":"2019-11-22T13:34:50Z","abstract_excerpt":"Learning models with discrete latent variables using stochastic gradient descent remains a challenge due to the high variance of gradient estimates. Modern variance reduction techniques mostly consider categorical distributions and have limited applicability when the number of possible outcomes becomes large. In this work, we consider models with latent permutations and propose control variates for the Plackett-Luce distribution. In particular, the control variates allow us to optimize black-box functions over permutations using stochastic gradient descent. To illustrate the approach, we consi"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1911.10036","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/1911.10036/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":"1911.10036","created_at":"2026-07-05T00:21:19.939432+00:00"},{"alias_kind":"arxiv_version","alias_value":"1911.10036v1","created_at":"2026-07-05T00:21:19.939432+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1911.10036","created_at":"2026-07-05T00:21:19.939432+00:00"},{"alias_kind":"pith_short_12","alias_value":"AO3EY7PSAQ6L","created_at":"2026-07-05T00:21:19.939432+00:00"},{"alias_kind":"pith_short_16","alias_value":"AO3EY7PSAQ6L55OG","created_at":"2026-07-05T00:21:19.939432+00:00"},{"alias_kind":"pith_short_8","alias_value":"AO3EY7PS","created_at":"2026-07-05T00:21:19.939432+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.26369","citing_title":"Scoring Is Not Enough: Addressing Gaps in Utility-fairness Trade-offs for Ranking","ref_index":20,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/AO3EY7PSAQ6L55OGX3GZK4KOFS","json":"https://pith.science/pith/AO3EY7PSAQ6L55OGX3GZK4KOFS.json","graph_json":"https://pith.science/api/pith-number/AO3EY7PSAQ6L55OGX3GZK4KOFS/graph.json","events_json":"https://pith.science/api/pith-number/AO3EY7PSAQ6L55OGX3GZK4KOFS/events.json","paper":"https://pith.science/paper/AO3EY7PS"},"agent_actions":{"view_html":"https://pith.science/pith/AO3EY7PSAQ6L55OGX3GZK4KOFS","download_json":"https://pith.science/pith/AO3EY7PSAQ6L55OGX3GZK4KOFS.json","view_paper":"https://pith.science/paper/AO3EY7PS","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1911.10036&json=true","fetch_graph":"https://pith.science/api/pith-number/AO3EY7PSAQ6L55OGX3GZK4KOFS/graph.json","fetch_events":"https://pith.science/api/pith-number/AO3EY7PSAQ6L55OGX3GZK4KOFS/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/AO3EY7PSAQ6L55OGX3GZK4KOFS/action/timestamp_anchor","attest_storage":"https://pith.science/pith/AO3EY7PSAQ6L55OGX3GZK4KOFS/action/storage_attestation","attest_author":"https://pith.science/pith/AO3EY7PSAQ6L55OGX3GZK4KOFS/action/author_attestation","sign_citation":"https://pith.science/pith/AO3EY7PSAQ6L55OGX3GZK4KOFS/action/citation_signature","submit_replication":"https://pith.science/pith/AO3EY7PSAQ6L55OGX3GZK4KOFS/action/replication_record"}},"created_at":"2026-07-05T00:21:19.939432+00:00","updated_at":"2026-07-05T00:21:19.939432+00:00"}