{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2020:KPUNA7TFVRFKPF23TIZMMQGAQB","short_pith_number":"pith:KPUNA7TF","schema_version":"1.0","canonical_sha256":"53e8d07e65ac4aa7975b9a32c640c0806c90186b9c9b65269281977062dc660c","source":{"kind":"arxiv","id":"2003.04125","version":1},"attestation_state":"computed","paper":{"title":"Amortized variance reduction for doubly stochastic objectives","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","stat.ME"],"primary_cat":"stat.ML","authors_text":"Ayman Boustati, James Hensman, Sattar Vakili, ST John","submitted_at":"2020-03-09T13:23:14Z","abstract_excerpt":"Approximate inference in complex probabilistic models such as deep Gaussian processes requires the optimisation of doubly stochastic objective functions. These objectives incorporate randomness both from mini-batch subsampling of the data and from Monte Carlo estimation of expectations. If the gradient variance is high, the stochastic optimisation problem becomes difficult with a slow rate of convergence. Control variates can be used to reduce the variance, but past approaches do not take into account how mini-batch stochasticity affects sampling stochasticity, resulting in sub-optimal varianc"},"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":"2003.04125","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"stat.ML","submitted_at":"2020-03-09T13:23:14Z","cross_cats_sorted":["cs.LG","stat.ME"],"title_canon_sha256":"e5c1eb62567db369fbe8f29640b02ed4ba3cb0b8bc8407393928ee0b73e38f34","abstract_canon_sha256":"dbd2f9fe2fe2a181fcd102b3293181c25fbb6389772ee9dbd10403770290b58f"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T00:50:29.089817Z","signature_b64":"5JdtOnOVYptoM8Cqjlq8zk59+iDJzozEIBny+KtY/7YkvGEKAfbZDobcAvIPhhI9zPgjXuNIRQyW9bYf4Gq7Bg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"53e8d07e65ac4aa7975b9a32c640c0806c90186b9c9b65269281977062dc660c","last_reissued_at":"2026-07-05T00:50:29.089388Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T00:50:29.089388Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Amortized variance reduction for doubly stochastic objectives","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","stat.ME"],"primary_cat":"stat.ML","authors_text":"Ayman Boustati, James Hensman, Sattar Vakili, ST John","submitted_at":"2020-03-09T13:23:14Z","abstract_excerpt":"Approximate inference in complex probabilistic models such as deep Gaussian processes requires the optimisation of doubly stochastic objective functions. These objectives incorporate randomness both from mini-batch subsampling of the data and from Monte Carlo estimation of expectations. If the gradient variance is high, the stochastic optimisation problem becomes difficult with a slow rate of convergence. Control variates can be used to reduce the variance, but past approaches do not take into account how mini-batch stochasticity affects sampling stochasticity, resulting in sub-optimal varianc"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2003.04125","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/2003.04125/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":"2003.04125","created_at":"2026-07-05T00:50:29.089448+00:00"},{"alias_kind":"arxiv_version","alias_value":"2003.04125v1","created_at":"2026-07-05T00:50:29.089448+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2003.04125","created_at":"2026-07-05T00:50:29.089448+00:00"},{"alias_kind":"pith_short_12","alias_value":"KPUNA7TFVRFK","created_at":"2026-07-05T00:50:29.089448+00:00"},{"alias_kind":"pith_short_16","alias_value":"KPUNA7TFVRFKPF23","created_at":"2026-07-05T00:50:29.089448+00:00"},{"alias_kind":"pith_short_8","alias_value":"KPUNA7TF","created_at":"2026-07-05T00:50:29.089448+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/KPUNA7TFVRFKPF23TIZMMQGAQB","json":"https://pith.science/pith/KPUNA7TFVRFKPF23TIZMMQGAQB.json","graph_json":"https://pith.science/api/pith-number/KPUNA7TFVRFKPF23TIZMMQGAQB/graph.json","events_json":"https://pith.science/api/pith-number/KPUNA7TFVRFKPF23TIZMMQGAQB/events.json","paper":"https://pith.science/paper/KPUNA7TF"},"agent_actions":{"view_html":"https://pith.science/pith/KPUNA7TFVRFKPF23TIZMMQGAQB","download_json":"https://pith.science/pith/KPUNA7TFVRFKPF23TIZMMQGAQB.json","view_paper":"https://pith.science/paper/KPUNA7TF","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2003.04125&json=true","fetch_graph":"https://pith.science/api/pith-number/KPUNA7TFVRFKPF23TIZMMQGAQB/graph.json","fetch_events":"https://pith.science/api/pith-number/KPUNA7TFVRFKPF23TIZMMQGAQB/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/KPUNA7TFVRFKPF23TIZMMQGAQB/action/timestamp_anchor","attest_storage":"https://pith.science/pith/KPUNA7TFVRFKPF23TIZMMQGAQB/action/storage_attestation","attest_author":"https://pith.science/pith/KPUNA7TFVRFKPF23TIZMMQGAQB/action/author_attestation","sign_citation":"https://pith.science/pith/KPUNA7TFVRFKPF23TIZMMQGAQB/action/citation_signature","submit_replication":"https://pith.science/pith/KPUNA7TFVRFKPF23TIZMMQGAQB/action/replication_record"}},"created_at":"2026-07-05T00:50:29.089448+00:00","updated_at":"2026-07-05T00:50:29.089448+00:00"}