{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:ZYWMUUN2PCLEBPC32JAHR2YPHS","short_pith_number":"pith:ZYWMUUN2","schema_version":"1.0","canonical_sha256":"ce2cca51ba789640bc5bd24078eb0f3cb8581ea74750ab9c9e1fe3878957b8a6","source":{"kind":"arxiv","id":"2501.03142","version":1},"attestation_state":"computed","paper":{"title":"Co-Activation Graph Analysis of Safety-Verified and Explainable Deep Reinforcement Learning Policies","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.AI","authors_text":"Dennis Gross, Helge Spieker","submitted_at":"2025-01-06T17:07:44Z","abstract_excerpt":"Deep reinforcement learning (RL) policies can demonstrate unsafe behaviors and are challenging to interpret. To address these challenges, we combine RL policy model checking--a technique for determining whether RL policies exhibit unsafe behaviors--with co-activation graph analysis--a method that maps neural network inner workings by analyzing neuron activation patterns--to gain insight into the safe RL policy's sequential decision-making. This combination lets us interpret the RL policy's inner workings for safe decision-making. We demonstrate its applicability in various experiments."},"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":"2501.03142","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-01-06T17:07:44Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"6397cd4dc1a44ab5c06bf1b4d3c06963e79d20c488e36147d77f1640267b6d7e","abstract_canon_sha256":"944b7542e6e352d2b6755e6ce0dc0cfef3322b99d6434b5bbd1cb641a44c9299"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:57:32.261205Z","signature_b64":"bTW6fwSsLs3I5SnEbLhHYPjQ9gXm2/BWOb0TwAQ45BhZPil4hmEdPCC5BIfWLUSMW4wuNGSlReMlKd5rum9mDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ce2cca51ba789640bc5bd24078eb0f3cb8581ea74750ab9c9e1fe3878957b8a6","last_reissued_at":"2026-07-05T09:57:32.260816Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:57:32.260816Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Co-Activation Graph Analysis of Safety-Verified and Explainable Deep Reinforcement Learning Policies","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.AI","authors_text":"Dennis Gross, Helge Spieker","submitted_at":"2025-01-06T17:07:44Z","abstract_excerpt":"Deep reinforcement learning (RL) policies can demonstrate unsafe behaviors and are challenging to interpret. To address these challenges, we combine RL policy model checking--a technique for determining whether RL policies exhibit unsafe behaviors--with co-activation graph analysis--a method that maps neural network inner workings by analyzing neuron activation patterns--to gain insight into the safe RL policy's sequential decision-making. This combination lets us interpret the RL policy's inner workings for safe decision-making. We demonstrate its applicability in various experiments."},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2501.03142","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/2501.03142/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":"2501.03142","created_at":"2026-07-05T09:57:32.260871+00:00"},{"alias_kind":"arxiv_version","alias_value":"2501.03142v1","created_at":"2026-07-05T09:57:32.260871+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2501.03142","created_at":"2026-07-05T09:57:32.260871+00:00"},{"alias_kind":"pith_short_12","alias_value":"ZYWMUUN2PCLE","created_at":"2026-07-05T09:57:32.260871+00:00"},{"alias_kind":"pith_short_16","alias_value":"ZYWMUUN2PCLEBPC3","created_at":"2026-07-05T09:57:32.260871+00:00"},{"alias_kind":"pith_short_8","alias_value":"ZYWMUUN2","created_at":"2026-07-05T09:57:32.260871+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/ZYWMUUN2PCLEBPC32JAHR2YPHS","json":"https://pith.science/pith/ZYWMUUN2PCLEBPC32JAHR2YPHS.json","graph_json":"https://pith.science/api/pith-number/ZYWMUUN2PCLEBPC32JAHR2YPHS/graph.json","events_json":"https://pith.science/api/pith-number/ZYWMUUN2PCLEBPC32JAHR2YPHS/events.json","paper":"https://pith.science/paper/ZYWMUUN2"},"agent_actions":{"view_html":"https://pith.science/pith/ZYWMUUN2PCLEBPC32JAHR2YPHS","download_json":"https://pith.science/pith/ZYWMUUN2PCLEBPC32JAHR2YPHS.json","view_paper":"https://pith.science/paper/ZYWMUUN2","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2501.03142&json=true","fetch_graph":"https://pith.science/api/pith-number/ZYWMUUN2PCLEBPC32JAHR2YPHS/graph.json","fetch_events":"https://pith.science/api/pith-number/ZYWMUUN2PCLEBPC32JAHR2YPHS/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ZYWMUUN2PCLEBPC32JAHR2YPHS/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ZYWMUUN2PCLEBPC32JAHR2YPHS/action/storage_attestation","attest_author":"https://pith.science/pith/ZYWMUUN2PCLEBPC32JAHR2YPHS/action/author_attestation","sign_citation":"https://pith.science/pith/ZYWMUUN2PCLEBPC32JAHR2YPHS/action/citation_signature","submit_replication":"https://pith.science/pith/ZYWMUUN2PCLEBPC32JAHR2YPHS/action/replication_record"}},"created_at":"2026-07-05T09:57:32.260871+00:00","updated_at":"2026-07-05T09:57:32.260871+00:00"}