{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2019:2QO5BPVWTVCK5WPIUKJZKDLFHK","short_pith_number":"pith:2QO5BPVW","schema_version":"1.0","canonical_sha256":"d41dd0beb69d44aed9e8a293950d653a8ccb8958fa3aa5a721efb94eeaf6e5be","source":{"kind":"arxiv","id":"1902.07198","version":4},"attestation_state":"computed","paper":{"title":"Learning to Generalize from Sparse and Underspecified Rewards","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CL","stat.ML"],"primary_cat":"cs.LG","authors_text":"Chen Liang, Dale Schuurmans, Mohammad Norouzi, Rishabh Agarwal","submitted_at":"2019-02-19T18:51:10Z","abstract_excerpt":"We consider the problem of learning from sparse and underspecified rewards, where an agent receives a complex input, such as a natural language instruction, and needs to generate a complex response, such as an action sequence, while only receiving binary success-failure feedback. Such success-failure rewards are often underspecified: they do not distinguish between purposeful and accidental success. Generalization from underspecified rewards hinges on discounting spurious trajectories that attain accidental success, while learning from sparse feedback requires effective exploration. We address"},"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":"1902.07198","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-02-19T18:51:10Z","cross_cats_sorted":["cs.AI","cs.CL","stat.ML"],"title_canon_sha256":"601869727f50719e7afd96c96db2715d098edb5602a20120ae4839eb9d1f22ab","abstract_canon_sha256":"6bd09c8e1badfddd1ecc401c4506db208945f2034be3a4051d91929cfbc9a50b"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T01:11:39.367484Z","signature_b64":"Ly6QX5+1DnK0kL0e031hyoxgnk8Qtl3lvdEtHcFAnvgo8HIgVPqWAu2uoV90VGTEB4vmr4D4FipL7Rj8P44dBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d41dd0beb69d44aed9e8a293950d653a8ccb8958fa3aa5a721efb94eeaf6e5be","last_reissued_at":"2026-07-05T01:11:39.367104Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T01:11:39.367104Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Learning to Generalize from Sparse and Underspecified Rewards","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CL","stat.ML"],"primary_cat":"cs.LG","authors_text":"Chen Liang, Dale Schuurmans, Mohammad Norouzi, Rishabh Agarwal","submitted_at":"2019-02-19T18:51:10Z","abstract_excerpt":"We consider the problem of learning from sparse and underspecified rewards, where an agent receives a complex input, such as a natural language instruction, and needs to generate a complex response, such as an action sequence, while only receiving binary success-failure feedback. Such success-failure rewards are often underspecified: they do not distinguish between purposeful and accidental success. Generalization from underspecified rewards hinges on discounting spurious trajectories that attain accidental success, while learning from sparse feedback requires effective exploration. We address"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1902.07198","kind":"arxiv","version":4},"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/1902.07198/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":"1902.07198","created_at":"2026-07-05T01:11:39.367171+00:00"},{"alias_kind":"arxiv_version","alias_value":"1902.07198v4","created_at":"2026-07-05T01:11:39.367171+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1902.07198","created_at":"2026-07-05T01:11:39.367171+00:00"},{"alias_kind":"pith_short_12","alias_value":"2QO5BPVWTVCK","created_at":"2026-07-05T01:11:39.367171+00:00"},{"alias_kind":"pith_short_16","alias_value":"2QO5BPVWTVCK5WPI","created_at":"2026-07-05T01:11:39.367171+00:00"},{"alias_kind":"pith_short_8","alias_value":"2QO5BPVW","created_at":"2026-07-05T01:11:39.367171+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"1908.11787","citing_title":"Answering Conversational Questions on Structured Data without Logical Forms","ref_index":3,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/2QO5BPVWTVCK5WPIUKJZKDLFHK","json":"https://pith.science/pith/2QO5BPVWTVCK5WPIUKJZKDLFHK.json","graph_json":"https://pith.science/api/pith-number/2QO5BPVWTVCK5WPIUKJZKDLFHK/graph.json","events_json":"https://pith.science/api/pith-number/2QO5BPVWTVCK5WPIUKJZKDLFHK/events.json","paper":"https://pith.science/paper/2QO5BPVW"},"agent_actions":{"view_html":"https://pith.science/pith/2QO5BPVWTVCK5WPIUKJZKDLFHK","download_json":"https://pith.science/pith/2QO5BPVWTVCK5WPIUKJZKDLFHK.json","view_paper":"https://pith.science/paper/2QO5BPVW","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1902.07198&json=true","fetch_graph":"https://pith.science/api/pith-number/2QO5BPVWTVCK5WPIUKJZKDLFHK/graph.json","fetch_events":"https://pith.science/api/pith-number/2QO5BPVWTVCK5WPIUKJZKDLFHK/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/2QO5BPVWTVCK5WPIUKJZKDLFHK/action/timestamp_anchor","attest_storage":"https://pith.science/pith/2QO5BPVWTVCK5WPIUKJZKDLFHK/action/storage_attestation","attest_author":"https://pith.science/pith/2QO5BPVWTVCK5WPIUKJZKDLFHK/action/author_attestation","sign_citation":"https://pith.science/pith/2QO5BPVWTVCK5WPIUKJZKDLFHK/action/citation_signature","submit_replication":"https://pith.science/pith/2QO5BPVWTVCK5WPIUKJZKDLFHK/action/replication_record"}},"created_at":"2026-07-05T01:11:39.367171+00:00","updated_at":"2026-07-05T01:11:39.367171+00:00"}