{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:LNGMEML4EKV5SAZSXUO6X5RVFV","short_pith_number":"pith:LNGMEML4","schema_version":"1.0","canonical_sha256":"5b4cc2317c22abd90332bd1debf6352d547e81365fb13c0e38fade6b9d10b35d","source":{"kind":"arxiv","id":"2212.12743","version":1},"attestation_state":"computed","paper":{"title":"Automated Gadget Discovery in Science","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"quant-ph","authors_text":"Andrea L\\'opez Incera, Fulvio Flamini, Hans J. Briegel, Hendrik Poulsen Nautrup, Lea M. Trenkwalder","submitted_at":"2022-12-24T14:52:22Z","abstract_excerpt":"In recent years, reinforcement learning (RL) has become increasingly successful in its application to science and the process of scientific discovery in general. However, while RL algorithms learn to solve increasingly complex problems, interpreting the solutions they provide becomes ever more challenging. In this work, we gain insights into an RL agent's learned behavior through a post-hoc analysis based on sequence mining and clustering. Specifically, frequent and compact subroutines, used by the agent to solve a given task, are distilled as gadgets and then grouped by various metrics. This "},"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":"2212.12743","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"quant-ph","submitted_at":"2022-12-24T14:52:22Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"a3e943296b3ec46cc4c557bebb40f77d6d5712fa4c47fd9d162470ddcd545de9","abstract_canon_sha256":"f18f16b53d19294051930827eec8ac46882e95851c6d39bd8db499f19d8f0ec6"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:28:12.534161Z","signature_b64":"d3TOP2Mq2JV1UkqUgu0sMKg+bUI/qtJlrZs4mmS4pVO9xq7aAnpoTkXuXQZ7vQb3QNnviJdDzyy2FE2IESCzBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"5b4cc2317c22abd90332bd1debf6352d547e81365fb13c0e38fade6b9d10b35d","last_reissued_at":"2026-07-05T05:28:12.533691Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:28:12.533691Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Automated Gadget Discovery in Science","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"quant-ph","authors_text":"Andrea L\\'opez Incera, Fulvio Flamini, Hans J. Briegel, Hendrik Poulsen Nautrup, Lea M. Trenkwalder","submitted_at":"2022-12-24T14:52:22Z","abstract_excerpt":"In recent years, reinforcement learning (RL) has become increasingly successful in its application to science and the process of scientific discovery in general. However, while RL algorithms learn to solve increasingly complex problems, interpreting the solutions they provide becomes ever more challenging. In this work, we gain insights into an RL agent's learned behavior through a post-hoc analysis based on sequence mining and clustering. Specifically, frequent and compact subroutines, used by the agent to solve a given task, are distilled as gadgets and then grouped by various metrics. This "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2212.12743","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/2212.12743/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":"2212.12743","created_at":"2026-07-05T05:28:12.533746+00:00"},{"alias_kind":"arxiv_version","alias_value":"2212.12743v1","created_at":"2026-07-05T05:28:12.533746+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2212.12743","created_at":"2026-07-05T05:28:12.533746+00:00"},{"alias_kind":"pith_short_12","alias_value":"LNGMEML4EKV5","created_at":"2026-07-05T05:28:12.533746+00:00"},{"alias_kind":"pith_short_16","alias_value":"LNGMEML4EKV5SAZS","created_at":"2026-07-05T05:28:12.533746+00:00"},{"alias_kind":"pith_short_8","alias_value":"LNGMEML4","created_at":"2026-07-05T05:28:12.533746+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2411.14991","citing_title":"Free Energy Projective Simulation (FEPS): Active inference with interpretability","ref_index":52,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/LNGMEML4EKV5SAZSXUO6X5RVFV","json":"https://pith.science/pith/LNGMEML4EKV5SAZSXUO6X5RVFV.json","graph_json":"https://pith.science/api/pith-number/LNGMEML4EKV5SAZSXUO6X5RVFV/graph.json","events_json":"https://pith.science/api/pith-number/LNGMEML4EKV5SAZSXUO6X5RVFV/events.json","paper":"https://pith.science/paper/LNGMEML4"},"agent_actions":{"view_html":"https://pith.science/pith/LNGMEML4EKV5SAZSXUO6X5RVFV","download_json":"https://pith.science/pith/LNGMEML4EKV5SAZSXUO6X5RVFV.json","view_paper":"https://pith.science/paper/LNGMEML4","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2212.12743&json=true","fetch_graph":"https://pith.science/api/pith-number/LNGMEML4EKV5SAZSXUO6X5RVFV/graph.json","fetch_events":"https://pith.science/api/pith-number/LNGMEML4EKV5SAZSXUO6X5RVFV/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/LNGMEML4EKV5SAZSXUO6X5RVFV/action/timestamp_anchor","attest_storage":"https://pith.science/pith/LNGMEML4EKV5SAZSXUO6X5RVFV/action/storage_attestation","attest_author":"https://pith.science/pith/LNGMEML4EKV5SAZSXUO6X5RVFV/action/author_attestation","sign_citation":"https://pith.science/pith/LNGMEML4EKV5SAZSXUO6X5RVFV/action/citation_signature","submit_replication":"https://pith.science/pith/LNGMEML4EKV5SAZSXUO6X5RVFV/action/replication_record"}},"created_at":"2026-07-05T05:28:12.533746+00:00","updated_at":"2026-07-05T05:28:12.533746+00:00"}