{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2020:QH3BTW33I2MALTU4D3XW5E2MBX","short_pith_number":"pith:QH3BTW33","schema_version":"1.0","canonical_sha256":"81f619db7b469805ce9c1eef6e934c0dd60d6297373256a95b4991d9981f16a7","source":{"kind":"arxiv","id":"2003.11881","version":2},"attestation_state":"computed","paper":{"title":"An empirical investigation of the challenges of real-world reinforcement learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Cosmin Paduraru, Daniel J. Mankowitz, Gabriel Dulac-Arnold, Jerry Li, Nir Levine, Sven Gowal, Todd Hester","submitted_at":"2020-03-24T11:05:41Z","abstract_excerpt":"Reinforcement learning (RL) has proven its worth in a series of artificial domains, and is beginning to show some successes in real-world scenarios. However, much of the research advances in RL are hard to leverage in real-world systems due to a series of assumptions that are rarely satisfied in practice. In this work, we identify and formalize a series of independent challenges that embody the difficulties that must be addressed for RL to be commonly deployed in real-world systems. For each challenge, we define it formally in the context of a Markov Decision Process, analyze the effects of th"},"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":"2003.11881","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2020-03-24T11:05:41Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"9a651c12ada03eae158b687d147384ce494a60314e739cf186b6fc665f4e14ce","abstract_canon_sha256":"4c29970f5083c314e848f4975e342eb025057adbe42279abd0cabf7aa52c1b1e"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:20:13.388966Z","signature_b64":"O8Mi7uMn6toQueJ29gicDXczZNjG0b3PFLWgOE/AN4GEb2YCDLgNo3X4bh58B0jNBN7xXj2LtMCArqkyC6B9BA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"81f619db7b469805ce9c1eef6e934c0dd60d6297373256a95b4991d9981f16a7","last_reissued_at":"2026-07-05T02:20:13.388473Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:20:13.388473Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"An empirical investigation of the challenges of real-world reinforcement learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Cosmin Paduraru, Daniel J. Mankowitz, Gabriel Dulac-Arnold, Jerry Li, Nir Levine, Sven Gowal, Todd Hester","submitted_at":"2020-03-24T11:05:41Z","abstract_excerpt":"Reinforcement learning (RL) has proven its worth in a series of artificial domains, and is beginning to show some successes in real-world scenarios. However, much of the research advances in RL are hard to leverage in real-world systems due to a series of assumptions that are rarely satisfied in practice. In this work, we identify and formalize a series of independent challenges that embody the difficulties that must be addressed for RL to be commonly deployed in real-world systems. For each challenge, we define it formally in the context of a Markov Decision Process, analyze the effects of th"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2003.11881","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/2003.11881/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":"2003.11881","created_at":"2026-07-05T02:20:13.388530+00:00"},{"alias_kind":"arxiv_version","alias_value":"2003.11881v2","created_at":"2026-07-05T02:20:13.388530+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2003.11881","created_at":"2026-07-05T02:20:13.388530+00:00"},{"alias_kind":"pith_short_12","alias_value":"QH3BTW33I2MA","created_at":"2026-07-05T02:20:13.388530+00:00"},{"alias_kind":"pith_short_16","alias_value":"QH3BTW33I2MALTU4","created_at":"2026-07-05T02:20:13.388530+00:00"},{"alias_kind":"pith_short_8","alias_value":"QH3BTW33","created_at":"2026-07-05T02:20:13.388530+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":4,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.12890","citing_title":"Learning to Adapt: Representation-Based Reinforcement Learning for Multi-Task Skill Transfer","ref_index":5,"is_internal_anchor":false},{"citing_arxiv_id":"2605.28287","citing_title":"AtomComposer: Discovering Chemical Space from First Principles with Reinforcement Learning","ref_index":5,"is_internal_anchor":false},{"citing_arxiv_id":"2605.19029","citing_title":"Distributionally Robust Control via Stein Variational Inference for Contact-Rich Manipulation","ref_index":8,"is_internal_anchor":false},{"citing_arxiv_id":"2004.07219","citing_title":"D4RL: Datasets for Deep Data-Driven Reinforcement Learning","ref_index":5,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/QH3BTW33I2MALTU4D3XW5E2MBX","json":"https://pith.science/pith/QH3BTW33I2MALTU4D3XW5E2MBX.json","graph_json":"https://pith.science/api/pith-number/QH3BTW33I2MALTU4D3XW5E2MBX/graph.json","events_json":"https://pith.science/api/pith-number/QH3BTW33I2MALTU4D3XW5E2MBX/events.json","paper":"https://pith.science/paper/QH3BTW33"},"agent_actions":{"view_html":"https://pith.science/pith/QH3BTW33I2MALTU4D3XW5E2MBX","download_json":"https://pith.science/pith/QH3BTW33I2MALTU4D3XW5E2MBX.json","view_paper":"https://pith.science/paper/QH3BTW33","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2003.11881&json=true","fetch_graph":"https://pith.science/api/pith-number/QH3BTW33I2MALTU4D3XW5E2MBX/graph.json","fetch_events":"https://pith.science/api/pith-number/QH3BTW33I2MALTU4D3XW5E2MBX/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/QH3BTW33I2MALTU4D3XW5E2MBX/action/timestamp_anchor","attest_storage":"https://pith.science/pith/QH3BTW33I2MALTU4D3XW5E2MBX/action/storage_attestation","attest_author":"https://pith.science/pith/QH3BTW33I2MALTU4D3XW5E2MBX/action/author_attestation","sign_citation":"https://pith.science/pith/QH3BTW33I2MALTU4D3XW5E2MBX/action/citation_signature","submit_replication":"https://pith.science/pith/QH3BTW33I2MALTU4D3XW5E2MBX/action/replication_record"}},"created_at":"2026-07-05T02:20:13.388530+00:00","updated_at":"2026-07-05T02:20:13.388530+00:00"}