{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:A6PJFTMDLEZJCNAKJEVZDDE6BF","short_pith_number":"pith:A6PJFTMD","schema_version":"1.0","canonical_sha256":"079e92cd83593291340a492b918c9e094731af7c98f83f1a52fd19c41e6ebb88","source":{"kind":"arxiv","id":"2403.12309","version":2},"attestation_state":"computed","paper":{"title":"Reinforcement Learning from Delayed Observations via World Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Armin Karamzade, Kyungmin Kim, Montek Kalsi, Roy Fox","submitted_at":"2024-03-18T23:18:27Z","abstract_excerpt":"In standard reinforcement learning settings, agents typically assume immediate feedback about the effects of their actions after taking them. However, in practice, this assumption may not hold true due to physical constraints and can significantly impact the performance of learning algorithms. In this paper, we address observation delays in partially observable environments. We propose leveraging world models, which have shown success in integrating past observations and learning dynamics, to handle observation delays. By reducing delayed POMDPs to delayed MDPs with world models, our methods c"},"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":"2403.12309","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-03-18T23:18:27Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"962e36f2f668b69f0652aab203855fa3d5c460788f5c19d571003927d3cb84cc","abstract_canon_sha256":"0ad3100b0e806f63ecf65af51e9fd0ddab289aa8627e26a74bf3f97c6504defa"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:36:46.655853Z","signature_b64":"9i1K4AFHS8jRaoKNNQhISjUVNy72+v1aQvnAmpD6v9Rz6u67PlRhNV0R8eNVPj9V2yhVr8Z0cKM4ywtsjxrKBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"079e92cd83593291340a492b918c9e094731af7c98f83f1a52fd19c41e6ebb88","last_reissued_at":"2026-07-05T08:36:46.655268Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:36:46.655268Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Reinforcement Learning from Delayed Observations via World Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Armin Karamzade, Kyungmin Kim, Montek Kalsi, Roy Fox","submitted_at":"2024-03-18T23:18:27Z","abstract_excerpt":"In standard reinforcement learning settings, agents typically assume immediate feedback about the effects of their actions after taking them. However, in practice, this assumption may not hold true due to physical constraints and can significantly impact the performance of learning algorithms. In this paper, we address observation delays in partially observable environments. We propose leveraging world models, which have shown success in integrating past observations and learning dynamics, to handle observation delays. By reducing delayed POMDPs to delayed MDPs with world models, our methods c"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2403.12309","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/2403.12309/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":"2403.12309","created_at":"2026-07-05T08:36:46.655337+00:00"},{"alias_kind":"arxiv_version","alias_value":"2403.12309v2","created_at":"2026-07-05T08:36:46.655337+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2403.12309","created_at":"2026-07-05T08:36:46.655337+00:00"},{"alias_kind":"pith_short_12","alias_value":"A6PJFTMDLEZJ","created_at":"2026-07-05T08:36:46.655337+00:00"},{"alias_kind":"pith_short_16","alias_value":"A6PJFTMDLEZJCNAK","created_at":"2026-07-05T08:36:46.655337+00:00"},{"alias_kind":"pith_short_8","alias_value":"A6PJFTMD","created_at":"2026-07-05T08:36:46.655337+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2509.20869","citing_title":"Model-Based Reinforcement Learning under Random Observation Delays","ref_index":19,"is_internal_anchor":false},{"citing_arxiv_id":"2605.12261","citing_title":"Delay-Empowered Causal Hierarchical Reinforcement Learning","ref_index":5,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/A6PJFTMDLEZJCNAKJEVZDDE6BF","json":"https://pith.science/pith/A6PJFTMDLEZJCNAKJEVZDDE6BF.json","graph_json":"https://pith.science/api/pith-number/A6PJFTMDLEZJCNAKJEVZDDE6BF/graph.json","events_json":"https://pith.science/api/pith-number/A6PJFTMDLEZJCNAKJEVZDDE6BF/events.json","paper":"https://pith.science/paper/A6PJFTMD"},"agent_actions":{"view_html":"https://pith.science/pith/A6PJFTMDLEZJCNAKJEVZDDE6BF","download_json":"https://pith.science/pith/A6PJFTMDLEZJCNAKJEVZDDE6BF.json","view_paper":"https://pith.science/paper/A6PJFTMD","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2403.12309&json=true","fetch_graph":"https://pith.science/api/pith-number/A6PJFTMDLEZJCNAKJEVZDDE6BF/graph.json","fetch_events":"https://pith.science/api/pith-number/A6PJFTMDLEZJCNAKJEVZDDE6BF/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/A6PJFTMDLEZJCNAKJEVZDDE6BF/action/timestamp_anchor","attest_storage":"https://pith.science/pith/A6PJFTMDLEZJCNAKJEVZDDE6BF/action/storage_attestation","attest_author":"https://pith.science/pith/A6PJFTMDLEZJCNAKJEVZDDE6BF/action/author_attestation","sign_citation":"https://pith.science/pith/A6PJFTMDLEZJCNAKJEVZDDE6BF/action/citation_signature","submit_replication":"https://pith.science/pith/A6PJFTMDLEZJCNAKJEVZDDE6BF/action/replication_record"}},"created_at":"2026-07-05T08:36:46.655337+00:00","updated_at":"2026-07-05T08:36:46.655337+00:00"}