{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:XKCFFKPBUEEKBUB6UUMPIEMJ57","short_pith_number":"pith:XKCFFKPB","schema_version":"1.0","canonical_sha256":"ba8452a9e1a108a0d03ea518f41189efe16965c1454d42166d7f170e96ecffad","source":{"kind":"arxiv","id":"2103.09159","version":5},"attestation_state":"computed","paper":{"title":"Learning to Shape Rewards using a Game of Two Partners","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.GT"],"primary_cat":"cs.LG","authors_text":"David Mguni, Feifei Tong, Hui Chen, Jiangcheng Zhu, Jianhong Wang, Jun Wang, Matthew Taylor, Nicolas Perez-Nieves, Taher Jafferjee, Tianpei Yang, Wenbin Song, Yaodong Yang","submitted_at":"2021-03-16T15:56:57Z","abstract_excerpt":"Reward shaping (RS) is a powerful method in reinforcement learning (RL) for overcoming the problem of sparse or uninformative rewards. However, RS typically relies on manually engineered shaping-reward functions whose construction is time-consuming and error-prone. It also requires domain knowledge which runs contrary to the goal of autonomous learning. We introduce Reinforcement Learning Optimising Shaping Algorithm (ROSA), an automated reward shaping framework in which the shaping-reward function is constructed in a Markov game between two agents. A reward-shaping agent (Shaper) uses switchi"},"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":"2103.09159","kind":"arxiv","version":5},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2021-03-16T15:56:57Z","cross_cats_sorted":["cs.AI","cs.GT"],"title_canon_sha256":"cca190eade985719442fb86266c33819fc8c7bb825318a411516f881747a1357","abstract_canon_sha256":"ea9b082562a89b9ea9a6dc29b94444d4600e0173d7a93d8016249f5c8df48dc4"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:38:59.997626Z","signature_b64":"+tzP96WGnNk6QvPRIKf4JCgF1asYsMt5BCQv++GEqKCVzOCEfBGpq6MC4DLYyOROvUPHJWS0yz/TLAw0RVKrBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ba8452a9e1a108a0d03ea518f41189efe16965c1454d42166d7f170e96ecffad","last_reissued_at":"2026-07-05T05:38:59.997151Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:38:59.997151Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Learning to Shape Rewards using a Game of Two Partners","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.GT"],"primary_cat":"cs.LG","authors_text":"David Mguni, Feifei Tong, Hui Chen, Jiangcheng Zhu, Jianhong Wang, Jun Wang, Matthew Taylor, Nicolas Perez-Nieves, Taher Jafferjee, Tianpei Yang, Wenbin Song, Yaodong Yang","submitted_at":"2021-03-16T15:56:57Z","abstract_excerpt":"Reward shaping (RS) is a powerful method in reinforcement learning (RL) for overcoming the problem of sparse or uninformative rewards. However, RS typically relies on manually engineered shaping-reward functions whose construction is time-consuming and error-prone. It also requires domain knowledge which runs contrary to the goal of autonomous learning. We introduce Reinforcement Learning Optimising Shaping Algorithm (ROSA), an automated reward shaping framework in which the shaping-reward function is constructed in a Markov game between two agents. A reward-shaping agent (Shaper) uses switchi"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2103.09159","kind":"arxiv","version":5},"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/2103.09159/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":"2103.09159","created_at":"2026-07-05T05:38:59.997208+00:00"},{"alias_kind":"arxiv_version","alias_value":"2103.09159v5","created_at":"2026-07-05T05:38:59.997208+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2103.09159","created_at":"2026-07-05T05:38:59.997208+00:00"},{"alias_kind":"pith_short_12","alias_value":"XKCFFKPBUEEK","created_at":"2026-07-05T05:38:59.997208+00:00"},{"alias_kind":"pith_short_16","alias_value":"XKCFFKPBUEEKBUB6","created_at":"2026-07-05T05:38:59.997208+00:00"},{"alias_kind":"pith_short_8","alias_value":"XKCFFKPB","created_at":"2026-07-05T05:38:59.997208+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2604.20627","citing_title":"Occupancy Reward Shaping: Improving Credit Assignment for Offline Goal-Conditioned Reinforcement Learning","ref_index":16,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/XKCFFKPBUEEKBUB6UUMPIEMJ57","json":"https://pith.science/pith/XKCFFKPBUEEKBUB6UUMPIEMJ57.json","graph_json":"https://pith.science/api/pith-number/XKCFFKPBUEEKBUB6UUMPIEMJ57/graph.json","events_json":"https://pith.science/api/pith-number/XKCFFKPBUEEKBUB6UUMPIEMJ57/events.json","paper":"https://pith.science/paper/XKCFFKPB"},"agent_actions":{"view_html":"https://pith.science/pith/XKCFFKPBUEEKBUB6UUMPIEMJ57","download_json":"https://pith.science/pith/XKCFFKPBUEEKBUB6UUMPIEMJ57.json","view_paper":"https://pith.science/paper/XKCFFKPB","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2103.09159&json=true","fetch_graph":"https://pith.science/api/pith-number/XKCFFKPBUEEKBUB6UUMPIEMJ57/graph.json","fetch_events":"https://pith.science/api/pith-number/XKCFFKPBUEEKBUB6UUMPIEMJ57/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/XKCFFKPBUEEKBUB6UUMPIEMJ57/action/timestamp_anchor","attest_storage":"https://pith.science/pith/XKCFFKPBUEEKBUB6UUMPIEMJ57/action/storage_attestation","attest_author":"https://pith.science/pith/XKCFFKPBUEEKBUB6UUMPIEMJ57/action/author_attestation","sign_citation":"https://pith.science/pith/XKCFFKPBUEEKBUB6UUMPIEMJ57/action/citation_signature","submit_replication":"https://pith.science/pith/XKCFFKPBUEEKBUB6UUMPIEMJ57/action/replication_record"}},"created_at":"2026-07-05T05:38:59.997208+00:00","updated_at":"2026-07-05T05:38:59.997208+00:00"}