{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:BEGFIHKCFNJTDG5IFM45LPCOF4","short_pith_number":"pith:BEGFIHKC","schema_version":"1.0","canonical_sha256":"090c541d422b53319ba82b39d5bc4e2f190f779093d517081d4ed23fcb8d8940","source":{"kind":"arxiv","id":"2311.12970","version":2},"attestation_state":"computed","paper":{"title":"Clustered Policy Decision Ranking","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Hana Chockler, Mark Levin","submitted_at":"2023-11-21T20:16:02Z","abstract_excerpt":"Policies trained via reinforcement learning (RL) are often very complex even for simple tasks. In an episode with n time steps, a policy will make n decisions on actions to take, many of which may appear non-intuitive to the observer. Moreover, it is not clear which of these decisions directly contribute towards achieving the reward and how significant their contribution is. Given a trained policy, we propose a black-box method based on statistical covariance estimation that clusters the states of the environment and ranks each cluster according to the importance of decisions made in its state"},"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":"2311.12970","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2023-11-21T20:16:02Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"99e8ccc73cf997ddfb61161a8abd9444386dd394f49c579d74c1ea08b98d505e","abstract_canon_sha256":"72ec785f8fa9e52b9cc29ba74a1ee7a8966281c607d1d3ea4469ff18fcd6ac02"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:12:46.363158Z","signature_b64":"rcRQJ998Dvy9fFZ8z5iz84HxhDIOqyl1UfPu7VyGttEm7/7VW6IgXz6oE6MMseCcrcg8hytW3wCSYzhV4GcOAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"090c541d422b53319ba82b39d5bc4e2f190f779093d517081d4ed23fcb8d8940","last_reissued_at":"2026-07-05T08:12:46.362627Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:12:46.362627Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Clustered Policy Decision Ranking","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Hana Chockler, Mark Levin","submitted_at":"2023-11-21T20:16:02Z","abstract_excerpt":"Policies trained via reinforcement learning (RL) are often very complex even for simple tasks. In an episode with n time steps, a policy will make n decisions on actions to take, many of which may appear non-intuitive to the observer. Moreover, it is not clear which of these decisions directly contribute towards achieving the reward and how significant their contribution is. Given a trained policy, we propose a black-box method based on statistical covariance estimation that clusters the states of the environment and ranks each cluster according to the importance of decisions made in its state"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2311.12970","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/2311.12970/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":"2311.12970","created_at":"2026-07-05T08:12:46.362684+00:00"},{"alias_kind":"arxiv_version","alias_value":"2311.12970v2","created_at":"2026-07-05T08:12:46.362684+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2311.12970","created_at":"2026-07-05T08:12:46.362684+00:00"},{"alias_kind":"pith_short_12","alias_value":"BEGFIHKCFNJT","created_at":"2026-07-05T08:12:46.362684+00:00"},{"alias_kind":"pith_short_16","alias_value":"BEGFIHKCFNJTDG5I","created_at":"2026-07-05T08:12:46.362684+00:00"},{"alias_kind":"pith_short_8","alias_value":"BEGFIHKC","created_at":"2026-07-05T08:12:46.362684+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2502.00726","citing_title":"Perspectives for Direct Interpretability in Multi-Agent Deep Reinforcement Learning","ref_index":67,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/BEGFIHKCFNJTDG5IFM45LPCOF4","json":"https://pith.science/pith/BEGFIHKCFNJTDG5IFM45LPCOF4.json","graph_json":"https://pith.science/api/pith-number/BEGFIHKCFNJTDG5IFM45LPCOF4/graph.json","events_json":"https://pith.science/api/pith-number/BEGFIHKCFNJTDG5IFM45LPCOF4/events.json","paper":"https://pith.science/paper/BEGFIHKC"},"agent_actions":{"view_html":"https://pith.science/pith/BEGFIHKCFNJTDG5IFM45LPCOF4","download_json":"https://pith.science/pith/BEGFIHKCFNJTDG5IFM45LPCOF4.json","view_paper":"https://pith.science/paper/BEGFIHKC","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2311.12970&json=true","fetch_graph":"https://pith.science/api/pith-number/BEGFIHKCFNJTDG5IFM45LPCOF4/graph.json","fetch_events":"https://pith.science/api/pith-number/BEGFIHKCFNJTDG5IFM45LPCOF4/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/BEGFIHKCFNJTDG5IFM45LPCOF4/action/timestamp_anchor","attest_storage":"https://pith.science/pith/BEGFIHKCFNJTDG5IFM45LPCOF4/action/storage_attestation","attest_author":"https://pith.science/pith/BEGFIHKCFNJTDG5IFM45LPCOF4/action/author_attestation","sign_citation":"https://pith.science/pith/BEGFIHKCFNJTDG5IFM45LPCOF4/action/citation_signature","submit_replication":"https://pith.science/pith/BEGFIHKCFNJTDG5IFM45LPCOF4/action/replication_record"}},"created_at":"2026-07-05T08:12:46.362684+00:00","updated_at":"2026-07-05T08:12:46.362684+00:00"}