{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:74Z7J2GMZ7ETSAF6PJ3KHHSZZE","short_pith_number":"pith:74Z7J2GM","schema_version":"1.0","canonical_sha256":"ff33f4e8cccfc93900be7a76a39e59c9299920211c1578502600084124ac1d2b","source":{"kind":"arxiv","id":"2605.26657","version":1},"attestation_state":"computed","paper":{"title":"Completion vs Optimality: Policy Gradient in Long-Horizon Cumulative-Damage Problems","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Sabine Janzen, Wolfgang Maass","submitted_at":"2026-05-26T07:43:03Z","abstract_excerpt":"Long-horizon decision problems with cumulative damage couple locally attractive actions to globally adverse outcomes. We identify two orthogonal failure modes for policy-gradient methods on this class and propose a decomposition that separates them: \\emph{completion} (reaching the terminal horizon rather than exiting via an implicit terminal constraint) and \\emph{optimality} (matching the dynamic-programming reference given completion). Under PPO with a linear soft penalty, granting horizon access alone reduces the completion rate: the penalty's equilibrium drives the dominant-activity share t"},"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":"2605.26657","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-05-26T07:43:03Z","cross_cats_sorted":[],"title_canon_sha256":"f2e0bd4639d7d224bbc9dd8c6b78d263a2b1252edc988e84616725bbf570b2b3","abstract_canon_sha256":"4529e9c2b63395c8324c3123ea3abb9fc96fed42166eb3eb3d286912d55cbdd7"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-27T01:06:04.359239Z","signature_b64":"+LViSyGJ1XMl+t4QgcNR/ktr6ImNyNnJnsRA1Xgrl0jN5ZKaBDl5rQqi3zGhxt9+noDnXRde5Mt530dlHzOEDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ff33f4e8cccfc93900be7a76a39e59c9299920211c1578502600084124ac1d2b","last_reissued_at":"2026-05-27T01:06:04.358468Z","signature_status":"signed_v1","first_computed_at":"2026-05-27T01:06:04.358468Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Completion vs Optimality: Policy Gradient in Long-Horizon Cumulative-Damage Problems","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Sabine Janzen, Wolfgang Maass","submitted_at":"2026-05-26T07:43:03Z","abstract_excerpt":"Long-horizon decision problems with cumulative damage couple locally attractive actions to globally adverse outcomes. We identify two orthogonal failure modes for policy-gradient methods on this class and propose a decomposition that separates them: \\emph{completion} (reaching the terminal horizon rather than exiting via an implicit terminal constraint) and \\emph{optimality} (matching the dynamic-programming reference given completion). Under PPO with a linear soft penalty, granting horizon access alone reduces the completion rate: the penalty's equilibrium drives the dominant-activity share t"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.26657","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":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2605.26657/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":"2605.26657","created_at":"2026-05-27T01:06:04.358586+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.26657v1","created_at":"2026-05-27T01:06:04.358586+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.26657","created_at":"2026-05-27T01:06:04.358586+00:00"},{"alias_kind":"pith_short_12","alias_value":"74Z7J2GMZ7ET","created_at":"2026-05-27T01:06:04.358586+00:00"},{"alias_kind":"pith_short_16","alias_value":"74Z7J2GMZ7ETSAF6","created_at":"2026-05-27T01:06:04.358586+00:00"},{"alias_kind":"pith_short_8","alias_value":"74Z7J2GM","created_at":"2026-05-27T01:06:04.358586+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/74Z7J2GMZ7ETSAF6PJ3KHHSZZE","json":"https://pith.science/pith/74Z7J2GMZ7ETSAF6PJ3KHHSZZE.json","graph_json":"https://pith.science/api/pith-number/74Z7J2GMZ7ETSAF6PJ3KHHSZZE/graph.json","events_json":"https://pith.science/api/pith-number/74Z7J2GMZ7ETSAF6PJ3KHHSZZE/events.json","paper":"https://pith.science/paper/74Z7J2GM"},"agent_actions":{"view_html":"https://pith.science/pith/74Z7J2GMZ7ETSAF6PJ3KHHSZZE","download_json":"https://pith.science/pith/74Z7J2GMZ7ETSAF6PJ3KHHSZZE.json","view_paper":"https://pith.science/paper/74Z7J2GM","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.26657&json=true","fetch_graph":"https://pith.science/api/pith-number/74Z7J2GMZ7ETSAF6PJ3KHHSZZE/graph.json","fetch_events":"https://pith.science/api/pith-number/74Z7J2GMZ7ETSAF6PJ3KHHSZZE/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/74Z7J2GMZ7ETSAF6PJ3KHHSZZE/action/timestamp_anchor","attest_storage":"https://pith.science/pith/74Z7J2GMZ7ETSAF6PJ3KHHSZZE/action/storage_attestation","attest_author":"https://pith.science/pith/74Z7J2GMZ7ETSAF6PJ3KHHSZZE/action/author_attestation","sign_citation":"https://pith.science/pith/74Z7J2GMZ7ETSAF6PJ3KHHSZZE/action/citation_signature","submit_replication":"https://pith.science/pith/74Z7J2GMZ7ETSAF6PJ3KHHSZZE/action/replication_record"}},"created_at":"2026-05-27T01:06:04.358586+00:00","updated_at":"2026-05-27T01:06:04.358586+00:00"}