{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2019:TQGTHGPETTPOW7AGJ25TOY5J2T","short_pith_number":"pith:TQGTHGPE","schema_version":"1.0","canonical_sha256":"9c0d3399e49cdeeb7c064ebb3763a9d4d93b3929484f083603065b03cd98c640","source":{"kind":"arxiv","id":"1905.13320","version":1},"attestation_state":"computed","paper":{"title":"Combating the Compounding-Error Problem with a Multi-step Model","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","stat.ML"],"primary_cat":"cs.LG","authors_text":"Dipendra Misra, Kavosh Asadi, Michel L. Littman, Seungchan Kim","submitted_at":"2019-05-30T21:30:29Z","abstract_excerpt":"Model-based reinforcement learning is an appealing framework for creating agents that learn, plan, and act in sequential environments. Model-based algorithms typically involve learning a transition model that takes a state and an action and outputs the next state---a one-step model. This model can be composed with itself to enable predicting multiple steps into the future, but one-step prediction errors can get magnified, leading to unacceptable inaccuracy. This compounding-error problem plagues planning and undermines model-based reinforcement learning. In this paper, we address the compoundi"},"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":"1905.13320","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-05-30T21:30:29Z","cross_cats_sorted":["cs.AI","stat.ML"],"title_canon_sha256":"48cc4b6ef798e6417b1746df296da19d9866df3c0ebb2c1a4337339090ae9274","abstract_canon_sha256":"3c55ddc0c422efabf41c6644df036c052fa7ada4234118d5ddef62c9636f13a7"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:44:35.312564Z","signature_b64":"eQlF6cTDS1/yHoR8D+Fpw6zx+aa6ZWAZ9AoOddk7Ygv9ZdWKnSXNj1ypIy61s/HNAPDsxLG896+7YiTjbP+GDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9c0d3399e49cdeeb7c064ebb3763a9d4d93b3929484f083603065b03cd98c640","last_reissued_at":"2026-05-17T23:44:35.312101Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:44:35.312101Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Combating the Compounding-Error Problem with a Multi-step Model","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","stat.ML"],"primary_cat":"cs.LG","authors_text":"Dipendra Misra, Kavosh Asadi, Michel L. Littman, Seungchan Kim","submitted_at":"2019-05-30T21:30:29Z","abstract_excerpt":"Model-based reinforcement learning is an appealing framework for creating agents that learn, plan, and act in sequential environments. Model-based algorithms typically involve learning a transition model that takes a state and an action and outputs the next state---a one-step model. This model can be composed with itself to enable predicting multiple steps into the future, but one-step prediction errors can get magnified, leading to unacceptable inaccuracy. This compounding-error problem plagues planning and undermines model-based reinforcement learning. In this paper, we address the compoundi"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1905.13320","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":""},"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":"1905.13320","created_at":"2026-05-17T23:44:35.312175+00:00"},{"alias_kind":"arxiv_version","alias_value":"1905.13320v1","created_at":"2026-05-17T23:44:35.312175+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1905.13320","created_at":"2026-05-17T23:44:35.312175+00:00"},{"alias_kind":"pith_short_12","alias_value":"TQGTHGPETTPO","created_at":"2026-05-18T12:33:30.264802+00:00"},{"alias_kind":"pith_short_16","alias_value":"TQGTHGPETTPOW7AG","created_at":"2026-05-18T12:33:30.264802+00:00"},{"alias_kind":"pith_short_8","alias_value":"TQGTHGPE","created_at":"2026-05-18T12:33:30.264802+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":6,"internal_anchor_count":4,"sample":[{"citing_arxiv_id":"2605.04568","citing_title":"Dream-MPC: Gradient-Based Model Predictive Control with Latent Imagination","ref_index":1,"is_internal_anchor":true},{"citing_arxiv_id":"2605.17548","citing_title":"Rethinking Code Review in the Age of AI: A Vision for Agentic Code Review","ref_index":224,"is_internal_anchor":true},{"citing_arxiv_id":"2512.04341","citing_title":"Long-Horizon Model-Based Offline Reinforcement Learning Without Explicit Conservatism","ref_index":1,"is_internal_anchor":true},{"citing_arxiv_id":"2211.15657","citing_title":"Is Conditional Generative Modeling all you need for Decision-Making?","ref_index":126,"is_internal_anchor":true},{"citing_arxiv_id":"2605.04568","citing_title":"Dream-MPC: Gradient-Based Model Predictive Control with Latent Imagination","ref_index":1,"is_internal_anchor":false},{"citing_arxiv_id":"2604.09035","citing_title":"Advantage-Guided Diffusion for Model-Based Reinforcement Learning","ref_index":25,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/TQGTHGPETTPOW7AGJ25TOY5J2T","json":"https://pith.science/pith/TQGTHGPETTPOW7AGJ25TOY5J2T.json","graph_json":"https://pith.science/api/pith-number/TQGTHGPETTPOW7AGJ25TOY5J2T/graph.json","events_json":"https://pith.science/api/pith-number/TQGTHGPETTPOW7AGJ25TOY5J2T/events.json","paper":"https://pith.science/paper/TQGTHGPE"},"agent_actions":{"view_html":"https://pith.science/pith/TQGTHGPETTPOW7AGJ25TOY5J2T","download_json":"https://pith.science/pith/TQGTHGPETTPOW7AGJ25TOY5J2T.json","view_paper":"https://pith.science/paper/TQGTHGPE","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1905.13320&json=true","fetch_graph":"https://pith.science/api/pith-number/TQGTHGPETTPOW7AGJ25TOY5J2T/graph.json","fetch_events":"https://pith.science/api/pith-number/TQGTHGPETTPOW7AGJ25TOY5J2T/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/TQGTHGPETTPOW7AGJ25TOY5J2T/action/timestamp_anchor","attest_storage":"https://pith.science/pith/TQGTHGPETTPOW7AGJ25TOY5J2T/action/storage_attestation","attest_author":"https://pith.science/pith/TQGTHGPETTPOW7AGJ25TOY5J2T/action/author_attestation","sign_citation":"https://pith.science/pith/TQGTHGPETTPOW7AGJ25TOY5J2T/action/citation_signature","submit_replication":"https://pith.science/pith/TQGTHGPETTPOW7AGJ25TOY5J2T/action/replication_record"}},"created_at":"2026-05-17T23:44:35.312175+00:00","updated_at":"2026-05-17T23:44:35.312175+00:00"}