{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:476X5XX4DV472DP52KRTDFUI5J","short_pith_number":"pith:476X5XX4","schema_version":"1.0","canonical_sha256":"e7fd7edefc1d79fd0dfdd2a3319688ea525c84a3363f1e95e0bcec99232a2b13","source":{"kind":"arxiv","id":"2605.28516","version":1},"attestation_state":"computed","paper":{"title":"Conservative neural posterior estimation via distributionally robust training","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"stat.ML","authors_text":"Ayush Bharti, Charita Dellaporta, Fran\\c{c}ois-Xavier Briol, William Laplante, Yuga Hikida","submitted_at":"2026-05-27T14:17:22Z","abstract_excerpt":"Simulation-based inference with neural posterior estimation (NPE) often yields overconfident and unreliable posteriors under limited simulation budgets. To address this, we propose DRO-NPE, a distributionally robust approach that replaces the standard NPE objective with a worst-case loss over a Wasserstein ambiguity set. We introduce KL-based metrics for miscoverage and miscalibration, and use these to show that the DRO-NPE objective controls overfitting and reduces posterior overconfidence. Our method is tractable, parallelisable, and readily integrates with standard normalising flows. Across"},"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":"2605.28516","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"stat.ML","submitted_at":"2026-05-27T14:17:22Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"281a160b99e880bc3f416e52e35699a983303787c51e7887f30041880f44a549","abstract_canon_sha256":"7b05920e7a389119c3607ec87f0ffc44f59b61e65d810a7d5646c73d8be0307a"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-28T02:04:55.224548Z","signature_b64":"2pQZ4WfuNHq2A2Gi46AZaBgYs0Ol/WbL+BooJXEc755pY4I5DJRtebqQb9zy6ipyL8iiF2Fv2Tx1kXz4CxrQAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e7fd7edefc1d79fd0dfdd2a3319688ea525c84a3363f1e95e0bcec99232a2b13","last_reissued_at":"2026-05-28T02:04:55.224128Z","signature_status":"signed_v1","first_computed_at":"2026-05-28T02:04:55.224128Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Conservative neural posterior estimation via distributionally robust training","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"stat.ML","authors_text":"Ayush Bharti, Charita Dellaporta, Fran\\c{c}ois-Xavier Briol, William Laplante, Yuga Hikida","submitted_at":"2026-05-27T14:17:22Z","abstract_excerpt":"Simulation-based inference with neural posterior estimation (NPE) often yields overconfident and unreliable posteriors under limited simulation budgets. To address this, we propose DRO-NPE, a distributionally robust approach that replaces the standard NPE objective with a worst-case loss over a Wasserstein ambiguity set. We introduce KL-based metrics for miscoverage and miscalibration, and use these to show that the DRO-NPE objective controls overfitting and reduces posterior overconfidence. Our method is tractable, parallelisable, and readily integrates with standard normalising flows. Across"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.28516","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/2605.28516/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":"2605.28516","created_at":"2026-05-28T02:04:55.224194+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.28516v1","created_at":"2026-05-28T02:04:55.224194+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.28516","created_at":"2026-05-28T02:04:55.224194+00:00"},{"alias_kind":"pith_short_12","alias_value":"476X5XX4DV47","created_at":"2026-05-28T02:04:55.224194+00:00"},{"alias_kind":"pith_short_16","alias_value":"476X5XX4DV472DP5","created_at":"2026-05-28T02:04:55.224194+00:00"},{"alias_kind":"pith_short_8","alias_value":"476X5XX4","created_at":"2026-05-28T02:04:55.224194+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/476X5XX4DV472DP52KRTDFUI5J","json":"https://pith.science/pith/476X5XX4DV472DP52KRTDFUI5J.json","graph_json":"https://pith.science/api/pith-number/476X5XX4DV472DP52KRTDFUI5J/graph.json","events_json":"https://pith.science/api/pith-number/476X5XX4DV472DP52KRTDFUI5J/events.json","paper":"https://pith.science/paper/476X5XX4"},"agent_actions":{"view_html":"https://pith.science/pith/476X5XX4DV472DP52KRTDFUI5J","download_json":"https://pith.science/pith/476X5XX4DV472DP52KRTDFUI5J.json","view_paper":"https://pith.science/paper/476X5XX4","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.28516&json=true","fetch_graph":"https://pith.science/api/pith-number/476X5XX4DV472DP52KRTDFUI5J/graph.json","fetch_events":"https://pith.science/api/pith-number/476X5XX4DV472DP52KRTDFUI5J/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/476X5XX4DV472DP52KRTDFUI5J/action/timestamp_anchor","attest_storage":"https://pith.science/pith/476X5XX4DV472DP52KRTDFUI5J/action/storage_attestation","attest_author":"https://pith.science/pith/476X5XX4DV472DP52KRTDFUI5J/action/author_attestation","sign_citation":"https://pith.science/pith/476X5XX4DV472DP52KRTDFUI5J/action/citation_signature","submit_replication":"https://pith.science/pith/476X5XX4DV472DP52KRTDFUI5J/action/replication_record"}},"created_at":"2026-05-28T02:04:55.224194+00:00","updated_at":"2026-05-28T02:04:55.224194+00:00"}