{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:XUQEB7HO4WYGEG72UY6A3LIQWY","short_pith_number":"pith:XUQEB7HO","schema_version":"1.0","canonical_sha256":"bd2040fceee5b0621bfaa63c0dad10b602261c69612000e7f68fd8e3c3687b36","source":{"kind":"arxiv","id":"2304.11084","version":1},"attestation_state":"computed","paper":{"title":"Training Automated Defense Strategies Using Graph-based Cyber Attack Simulations","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG","cs.NI"],"primary_cat":"cs.CR","authors_text":"Jakob Nyberg, Pontus Johnson","submitted_at":"2023-04-17T07:52:00Z","abstract_excerpt":"We implemented and evaluated an automated cyber defense agent. The agent takes security alerts as input and uses reinforcement learning to learn a policy for executing predefined defensive measures. The defender policies were trained in an environment intended to simulate a cyber attack. In the simulation, an attacking agent attempts to capture targets in the environment, while the defender attempts to protect them by enabling defenses. The environment was modeled using attack graphs based on the Meta Attack Language language. We assumed that defensive measures have downtime costs, meaning tha"},"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":"2304.11084","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2023-04-17T07:52:00Z","cross_cats_sorted":["cs.LG","cs.NI"],"title_canon_sha256":"b88fe4289f7960be2e44300112020e9411a4c50aa67ac25fd75917d80d09e3bf","abstract_canon_sha256":"1ac29bea6c4b06dd7d260ca9babe71d5184ed5d10b0bd173ae60f3d16488128b"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:03:18.866496Z","signature_b64":"FTa9310pHaVhtAKzkQxM1agCRs4LN4oXRYiC6qnf5YBI/eOZNMFqTZcdgAZWd3/WD87NeS8Nh+cjYLsmNeLoAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"bd2040fceee5b0621bfaa63c0dad10b602261c69612000e7f68fd8e3c3687b36","last_reissued_at":"2026-07-05T06:03:18.865993Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:03:18.865993Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Training Automated Defense Strategies Using Graph-based Cyber Attack Simulations","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG","cs.NI"],"primary_cat":"cs.CR","authors_text":"Jakob Nyberg, Pontus Johnson","submitted_at":"2023-04-17T07:52:00Z","abstract_excerpt":"We implemented and evaluated an automated cyber defense agent. The agent takes security alerts as input and uses reinforcement learning to learn a policy for executing predefined defensive measures. The defender policies were trained in an environment intended to simulate a cyber attack. In the simulation, an attacking agent attempts to capture targets in the environment, while the defender attempts to protect them by enabling defenses. The environment was modeled using attack graphs based on the Meta Attack Language language. We assumed that defensive measures have downtime costs, meaning tha"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2304.11084","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/2304.11084/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":"2304.11084","created_at":"2026-07-05T06:03:18.866061+00:00"},{"alias_kind":"arxiv_version","alias_value":"2304.11084v1","created_at":"2026-07-05T06:03:18.866061+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2304.11084","created_at":"2026-07-05T06:03:18.866061+00:00"},{"alias_kind":"pith_short_12","alias_value":"XUQEB7HO4WYG","created_at":"2026-07-05T06:03:18.866061+00:00"},{"alias_kind":"pith_short_16","alias_value":"XUQEB7HO4WYGEG72","created_at":"2026-07-05T06:03:18.866061+00:00"},{"alias_kind":"pith_short_8","alias_value":"XUQEB7HO","created_at":"2026-07-05T06:03:18.866061+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2505.22531","citing_title":"Training RL Agents for Multi-Objective Network Defense Tasks","ref_index":30,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/XUQEB7HO4WYGEG72UY6A3LIQWY","json":"https://pith.science/pith/XUQEB7HO4WYGEG72UY6A3LIQWY.json","graph_json":"https://pith.science/api/pith-number/XUQEB7HO4WYGEG72UY6A3LIQWY/graph.json","events_json":"https://pith.science/api/pith-number/XUQEB7HO4WYGEG72UY6A3LIQWY/events.json","paper":"https://pith.science/paper/XUQEB7HO"},"agent_actions":{"view_html":"https://pith.science/pith/XUQEB7HO4WYGEG72UY6A3LIQWY","download_json":"https://pith.science/pith/XUQEB7HO4WYGEG72UY6A3LIQWY.json","view_paper":"https://pith.science/paper/XUQEB7HO","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2304.11084&json=true","fetch_graph":"https://pith.science/api/pith-number/XUQEB7HO4WYGEG72UY6A3LIQWY/graph.json","fetch_events":"https://pith.science/api/pith-number/XUQEB7HO4WYGEG72UY6A3LIQWY/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/XUQEB7HO4WYGEG72UY6A3LIQWY/action/timestamp_anchor","attest_storage":"https://pith.science/pith/XUQEB7HO4WYGEG72UY6A3LIQWY/action/storage_attestation","attest_author":"https://pith.science/pith/XUQEB7HO4WYGEG72UY6A3LIQWY/action/author_attestation","sign_citation":"https://pith.science/pith/XUQEB7HO4WYGEG72UY6A3LIQWY/action/citation_signature","submit_replication":"https://pith.science/pith/XUQEB7HO4WYGEG72UY6A3LIQWY/action/replication_record"}},"created_at":"2026-07-05T06:03:18.866061+00:00","updated_at":"2026-07-05T06:03:18.866061+00:00"}