{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:GQ5HVHTX5SCZP7EVHAAK5I65O6","short_pith_number":"pith:GQ5HVHTX","schema_version":"1.0","canonical_sha256":"343a7a9e77ec8597fc953800aea3dd779ea21f9a73cd27e69bb3d22dd6e5c82c","source":{"kind":"arxiv","id":"2107.06353","version":6},"attestation_state":"computed","paper":{"title":"Distributionally Robust Policy Learning via Adversarial Environment Generation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.RO","authors_text":"Allen Z. Ren, Anirudha Majumdar","submitted_at":"2021-07-13T19:26:34Z","abstract_excerpt":"Our goal is to train control policies that generalize well to unseen environments. Inspired by the Distributionally Robust Optimization (DRO) framework, we propose DRAGEN - Distributionally Robust policy learning via Adversarial Generation of ENvironments - for iteratively improving robustness of policies to realistic distribution shifts by generating adversarial environments. The key idea is to learn a generative model for environments whose latent variables capture cost-predictive and realistic variations in environments. We perform DRO with respect to a Wasserstein ball around the empirical"},"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":"2107.06353","kind":"arxiv","version":6},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.RO","submitted_at":"2021-07-13T19:26:34Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"143e29ef06f7b0dfad670f94ec7d52ae7f51c8adfdbc9290e9f2dc04e04e7e95","abstract_canon_sha256":"65ebd56653954b928575bff2caf91e8bf1553f8e379693ebdb14498b64d06d6c"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:38:11.044598Z","signature_b64":"i+gJru0547caaNs1mp3ysiIFpJuG6bSTKAG4znrpWBcMJpBaxgatJEutm2k/7ESHLCmBAOiY9CsHfw/9w5xQCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"343a7a9e77ec8597fc953800aea3dd779ea21f9a73cd27e69bb3d22dd6e5c82c","last_reissued_at":"2026-07-05T04:38:11.044139Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:38:11.044139Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Distributionally Robust Policy Learning via Adversarial Environment Generation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.RO","authors_text":"Allen Z. Ren, Anirudha Majumdar","submitted_at":"2021-07-13T19:26:34Z","abstract_excerpt":"Our goal is to train control policies that generalize well to unseen environments. Inspired by the Distributionally Robust Optimization (DRO) framework, we propose DRAGEN - Distributionally Robust policy learning via Adversarial Generation of ENvironments - for iteratively improving robustness of policies to realistic distribution shifts by generating adversarial environments. The key idea is to learn a generative model for environments whose latent variables capture cost-predictive and realistic variations in environments. We perform DRO with respect to a Wasserstein ball around the empirical"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2107.06353","kind":"arxiv","version":6},"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/2107.06353/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":"2107.06353","created_at":"2026-07-05T04:38:11.044202+00:00"},{"alias_kind":"arxiv_version","alias_value":"2107.06353v6","created_at":"2026-07-05T04:38:11.044202+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2107.06353","created_at":"2026-07-05T04:38:11.044202+00:00"},{"alias_kind":"pith_short_12","alias_value":"GQ5HVHTX5SCZ","created_at":"2026-07-05T04:38:11.044202+00:00"},{"alias_kind":"pith_short_16","alias_value":"GQ5HVHTX5SCZP7EV","created_at":"2026-07-05T04:38:11.044202+00:00"},{"alias_kind":"pith_short_8","alias_value":"GQ5HVHTX","created_at":"2026-07-05T04:38:11.044202+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2412.02818","citing_title":"RoboMD: Uncovering Robot Vulnerabilities through Semantic Potential Fields","ref_index":41,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/GQ5HVHTX5SCZP7EVHAAK5I65O6","json":"https://pith.science/pith/GQ5HVHTX5SCZP7EVHAAK5I65O6.json","graph_json":"https://pith.science/api/pith-number/GQ5HVHTX5SCZP7EVHAAK5I65O6/graph.json","events_json":"https://pith.science/api/pith-number/GQ5HVHTX5SCZP7EVHAAK5I65O6/events.json","paper":"https://pith.science/paper/GQ5HVHTX"},"agent_actions":{"view_html":"https://pith.science/pith/GQ5HVHTX5SCZP7EVHAAK5I65O6","download_json":"https://pith.science/pith/GQ5HVHTX5SCZP7EVHAAK5I65O6.json","view_paper":"https://pith.science/paper/GQ5HVHTX","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2107.06353&json=true","fetch_graph":"https://pith.science/api/pith-number/GQ5HVHTX5SCZP7EVHAAK5I65O6/graph.json","fetch_events":"https://pith.science/api/pith-number/GQ5HVHTX5SCZP7EVHAAK5I65O6/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/GQ5HVHTX5SCZP7EVHAAK5I65O6/action/timestamp_anchor","attest_storage":"https://pith.science/pith/GQ5HVHTX5SCZP7EVHAAK5I65O6/action/storage_attestation","attest_author":"https://pith.science/pith/GQ5HVHTX5SCZP7EVHAAK5I65O6/action/author_attestation","sign_citation":"https://pith.science/pith/GQ5HVHTX5SCZP7EVHAAK5I65O6/action/citation_signature","submit_replication":"https://pith.science/pith/GQ5HVHTX5SCZP7EVHAAK5I65O6/action/replication_record"}},"created_at":"2026-07-05T04:38:11.044202+00:00","updated_at":"2026-07-05T04:38:11.044202+00:00"}