{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:LUB73PYU4H6QRU2YG5GOOGOWLS","short_pith_number":"pith:LUB73PYU","schema_version":"1.0","canonical_sha256":"5d03fdbf14e1fd08d358374ce719d65c8c6d657062d806dc11d514efe31bf924","source":{"kind":"arxiv","id":"2106.14080","version":2},"attestation_state":"computed","paper":{"title":"Model-Advantage and Value-Aware Models for Model-Based Reinforcement Learning: Bridging the Gap in Theory and Practice","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","stat.ML"],"primary_cat":"cs.LG","authors_text":"Ashwin Kalyan, Dhruv Batra, Harish Kamath, Nirbhay Modhe","submitted_at":"2021-06-26T20:01:28Z","abstract_excerpt":"This work shows that value-aware model learning, known for its numerous theoretical benefits, is also practically viable for solving challenging continuous control tasks in prevalent model-based reinforcement learning algorithms. First, we derive a novel value-aware model learning objective by bounding the model-advantage i.e. model performance difference, between two MDPs or models given a fixed policy, achieving superior performance to prior value-aware objectives in most continuous control environments. Second, we identify the issue of stale value estimates in naively substituting value-awa"},"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":"2106.14080","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2021-06-26T20:01:28Z","cross_cats_sorted":["cs.AI","stat.ML"],"title_canon_sha256":"877120b574774f2b19da2b43b083ab58d11cc1f64fa3e9bc535ddb2b73511932","abstract_canon_sha256":"9a2eddb53cf2cedd03ffce6dcdf4975d2522192cdedc353ad44acf28f53d14ba"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:52:13.447630Z","signature_b64":"u+j+dzwfs+FVleBo3wSZensYIeM1smKOU0A3m5GsmJtT4cDlDAlk5XVJZkAglwjut+MQC9AX7LmrISto8JcgAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"5d03fdbf14e1fd08d358374ce719d65c8c6d657062d806dc11d514efe31bf924","last_reissued_at":"2026-07-05T03:52:13.447235Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:52:13.447235Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Model-Advantage and Value-Aware Models for Model-Based Reinforcement Learning: Bridging the Gap in Theory and Practice","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","stat.ML"],"primary_cat":"cs.LG","authors_text":"Ashwin Kalyan, Dhruv Batra, Harish Kamath, Nirbhay Modhe","submitted_at":"2021-06-26T20:01:28Z","abstract_excerpt":"This work shows that value-aware model learning, known for its numerous theoretical benefits, is also practically viable for solving challenging continuous control tasks in prevalent model-based reinforcement learning algorithms. First, we derive a novel value-aware model learning objective by bounding the model-advantage i.e. model performance difference, between two MDPs or models given a fixed policy, achieving superior performance to prior value-aware objectives in most continuous control environments. Second, we identify the issue of stale value estimates in naively substituting value-awa"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2106.14080","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/2106.14080/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":"2106.14080","created_at":"2026-07-05T03:52:13.447288+00:00"},{"alias_kind":"arxiv_version","alias_value":"2106.14080v2","created_at":"2026-07-05T03:52:13.447288+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2106.14080","created_at":"2026-07-05T03:52:13.447288+00:00"},{"alias_kind":"pith_short_12","alias_value":"LUB73PYU4H6Q","created_at":"2026-07-05T03:52:13.447288+00:00"},{"alias_kind":"pith_short_16","alias_value":"LUB73PYU4H6QRU2Y","created_at":"2026-07-05T03:52:13.447288+00:00"},{"alias_kind":"pith_short_8","alias_value":"LUB73PYU","created_at":"2026-07-05T03:52:13.447288+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/LUB73PYU4H6QRU2YG5GOOGOWLS","json":"https://pith.science/pith/LUB73PYU4H6QRU2YG5GOOGOWLS.json","graph_json":"https://pith.science/api/pith-number/LUB73PYU4H6QRU2YG5GOOGOWLS/graph.json","events_json":"https://pith.science/api/pith-number/LUB73PYU4H6QRU2YG5GOOGOWLS/events.json","paper":"https://pith.science/paper/LUB73PYU"},"agent_actions":{"view_html":"https://pith.science/pith/LUB73PYU4H6QRU2YG5GOOGOWLS","download_json":"https://pith.science/pith/LUB73PYU4H6QRU2YG5GOOGOWLS.json","view_paper":"https://pith.science/paper/LUB73PYU","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2106.14080&json=true","fetch_graph":"https://pith.science/api/pith-number/LUB73PYU4H6QRU2YG5GOOGOWLS/graph.json","fetch_events":"https://pith.science/api/pith-number/LUB73PYU4H6QRU2YG5GOOGOWLS/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/LUB73PYU4H6QRU2YG5GOOGOWLS/action/timestamp_anchor","attest_storage":"https://pith.science/pith/LUB73PYU4H6QRU2YG5GOOGOWLS/action/storage_attestation","attest_author":"https://pith.science/pith/LUB73PYU4H6QRU2YG5GOOGOWLS/action/author_attestation","sign_citation":"https://pith.science/pith/LUB73PYU4H6QRU2YG5GOOGOWLS/action/citation_signature","submit_replication":"https://pith.science/pith/LUB73PYU4H6QRU2YG5GOOGOWLS/action/replication_record"}},"created_at":"2026-07-05T03:52:13.447288+00:00","updated_at":"2026-07-05T03:52:13.447288+00:00"}