{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:V4EMCABF6SDTQFOG4WZCKOI75K","short_pith_number":"pith:V4EMCABF","schema_version":"1.0","canonical_sha256":"af08c10025f4873815c6e5b225391feab73c42d4e9d1d0ac5c500f60cd999f69","source":{"kind":"arxiv","id":"2305.18435","version":2},"attestation_state":"computed","paper":{"title":"Statistically Efficient Bayesian Sequential Experiment Design via Reinforcement Learning with Cross-Entropy Estimators","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["stat.ME"],"primary_cat":"cs.LG","authors_text":"Amir Dezfouli, Daniel Steinberg, Edwin V. Bonilla, Iadine Chades, Tom Blau","submitted_at":"2023-05-29T00:35:52Z","abstract_excerpt":"Reinforcement learning can learn amortised design policies for designing sequences of experiments. However, current amortised methods rely on estimators of expected information gain (EIG) that require an exponential number of samples on the magnitude of the EIG to achieve an unbiased estimation. We propose the use of an alternative estimator based on the cross-entropy of the joint model distribution and a flexible proposal distribution. This proposal distribution approximates the true posterior of the model parameters given the experimental history and the design policy. Our method overcomes t"},"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":"2305.18435","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.LG","submitted_at":"2023-05-29T00:35:52Z","cross_cats_sorted":["stat.ME"],"title_canon_sha256":"d7965c7cab8c7dcf4fafa661feb2837df4b5dace04e44accf64ec33c1da6268a","abstract_canon_sha256":"c0f2ed85128c8f2b044fa4f53e4e410ef4266f865e0ea6d9eb41b2035734a69a"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:40:53.789844Z","signature_b64":"sK3ZhUAuem3FAzZyC+PF8c/PfnCSRqW88aGMTitwFZykARJOH44rMtfbF9vkCFrPPLLfRnWZuuGKvqkoVCljDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"af08c10025f4873815c6e5b225391feab73c42d4e9d1d0ac5c500f60cd999f69","last_reissued_at":"2026-07-05T07:40:53.789353Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:40:53.789353Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Statistically Efficient Bayesian Sequential Experiment Design via Reinforcement Learning with Cross-Entropy Estimators","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["stat.ME"],"primary_cat":"cs.LG","authors_text":"Amir Dezfouli, Daniel Steinberg, Edwin V. Bonilla, Iadine Chades, Tom Blau","submitted_at":"2023-05-29T00:35:52Z","abstract_excerpt":"Reinforcement learning can learn amortised design policies for designing sequences of experiments. However, current amortised methods rely on estimators of expected information gain (EIG) that require an exponential number of samples on the magnitude of the EIG to achieve an unbiased estimation. We propose the use of an alternative estimator based on the cross-entropy of the joint model distribution and a flexible proposal distribution. This proposal distribution approximates the true posterior of the model parameters given the experimental history and the design policy. Our method overcomes t"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2305.18435","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/2305.18435/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":"2305.18435","created_at":"2026-07-05T07:40:53.789413+00:00"},{"alias_kind":"arxiv_version","alias_value":"2305.18435v2","created_at":"2026-07-05T07:40:53.789413+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2305.18435","created_at":"2026-07-05T07:40:53.789413+00:00"},{"alias_kind":"pith_short_12","alias_value":"V4EMCABF6SDT","created_at":"2026-07-05T07:40:53.789413+00:00"},{"alias_kind":"pith_short_16","alias_value":"V4EMCABF6SDTQFOG","created_at":"2026-07-05T07:40:53.789413+00:00"},{"alias_kind":"pith_short_8","alias_value":"V4EMCABF","created_at":"2026-07-05T07:40:53.789413+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.25197","citing_title":"Efficient Adaptive Data Acquisition via Pretrained Belief Representations","ref_index":6,"is_internal_anchor":false},{"citing_arxiv_id":"2605.26990","citing_title":"Constrained Bayesian Experimental Design via Online Planning","ref_index":1,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/V4EMCABF6SDTQFOG4WZCKOI75K","json":"https://pith.science/pith/V4EMCABF6SDTQFOG4WZCKOI75K.json","graph_json":"https://pith.science/api/pith-number/V4EMCABF6SDTQFOG4WZCKOI75K/graph.json","events_json":"https://pith.science/api/pith-number/V4EMCABF6SDTQFOG4WZCKOI75K/events.json","paper":"https://pith.science/paper/V4EMCABF"},"agent_actions":{"view_html":"https://pith.science/pith/V4EMCABF6SDTQFOG4WZCKOI75K","download_json":"https://pith.science/pith/V4EMCABF6SDTQFOG4WZCKOI75K.json","view_paper":"https://pith.science/paper/V4EMCABF","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2305.18435&json=true","fetch_graph":"https://pith.science/api/pith-number/V4EMCABF6SDTQFOG4WZCKOI75K/graph.json","fetch_events":"https://pith.science/api/pith-number/V4EMCABF6SDTQFOG4WZCKOI75K/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/V4EMCABF6SDTQFOG4WZCKOI75K/action/timestamp_anchor","attest_storage":"https://pith.science/pith/V4EMCABF6SDTQFOG4WZCKOI75K/action/storage_attestation","attest_author":"https://pith.science/pith/V4EMCABF6SDTQFOG4WZCKOI75K/action/author_attestation","sign_citation":"https://pith.science/pith/V4EMCABF6SDTQFOG4WZCKOI75K/action/citation_signature","submit_replication":"https://pith.science/pith/V4EMCABF6SDTQFOG4WZCKOI75K/action/replication_record"}},"created_at":"2026-07-05T07:40:53.789413+00:00","updated_at":"2026-07-05T07:40:53.789413+00:00"}