{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:27VNV4VTAZQJ4T224D3JFWZBHL","short_pith_number":"pith:27VNV4VT","schema_version":"1.0","canonical_sha256":"d7eadaf2b306609e4f5ae0f692db213ad4a2a72a1a969a3f74a3ff3c33164eaa","source":{"kind":"arxiv","id":"2501.19206","version":1},"attestation_state":"computed","paper":{"title":"An Empirical Game-Theoretic Analysis of Autonomous Cyber-Defence Agents","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.CR","cs.GT"],"primary_cat":"cs.AI","authors_text":"Alex Hiles, Chris Willis, Daniel J.B. Harrold, Gregory Palmer, Ian Miles, Luke Swaby, Matthew Stewart, Sara Farmer","submitted_at":"2025-01-31T15:15:02Z","abstract_excerpt":"The recent rise in increasingly sophisticated cyber-attacks raises the need for robust and resilient autonomous cyber-defence (ACD) agents. Given the variety of cyber-attack tactics, techniques and procedures (TTPs) employed, learning approaches that can return generalisable policies are desirable. Meanwhile, the assurance of ACD agents remains an open challenge. We address both challenges via an empirical game-theoretic analysis of deep reinforcement learning (DRL) approaches for ACD using the principled double oracle (DO) algorithm. This algorithm relies on adversaries iteratively learning ("},"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":"2501.19206","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.AI","submitted_at":"2025-01-31T15:15:02Z","cross_cats_sorted":["cs.CR","cs.GT"],"title_canon_sha256":"a11b069771855f6d2103399ebec4ace40d3339dd419c523d5a5f8d3dfd870209","abstract_canon_sha256":"31d2968f149847137109d38ef3e628308525b288f5e46c9d1478b90b9b58c503"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:07:57.755298Z","signature_b64":"LQKehD07RTPmzm7nWLiHPq8P5qTkzfrX5Mb0jCDG2lSmulAeFpelhqKt4K7qjIxeYI0zEV+qpPG3MaHrE+lWBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d7eadaf2b306609e4f5ae0f692db213ad4a2a72a1a969a3f74a3ff3c33164eaa","last_reissued_at":"2026-07-05T10:07:57.754741Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:07:57.754741Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"An Empirical Game-Theoretic Analysis of Autonomous Cyber-Defence Agents","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.CR","cs.GT"],"primary_cat":"cs.AI","authors_text":"Alex Hiles, Chris Willis, Daniel J.B. Harrold, Gregory Palmer, Ian Miles, Luke Swaby, Matthew Stewart, Sara Farmer","submitted_at":"2025-01-31T15:15:02Z","abstract_excerpt":"The recent rise in increasingly sophisticated cyber-attacks raises the need for robust and resilient autonomous cyber-defence (ACD) agents. Given the variety of cyber-attack tactics, techniques and procedures (TTPs) employed, learning approaches that can return generalisable policies are desirable. Meanwhile, the assurance of ACD agents remains an open challenge. We address both challenges via an empirical game-theoretic analysis of deep reinforcement learning (DRL) approaches for ACD using the principled double oracle (DO) algorithm. This algorithm relies on adversaries iteratively learning ("},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2501.19206","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/2501.19206/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":"2501.19206","created_at":"2026-07-05T10:07:57.754789+00:00"},{"alias_kind":"arxiv_version","alias_value":"2501.19206v1","created_at":"2026-07-05T10:07:57.754789+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2501.19206","created_at":"2026-07-05T10:07:57.754789+00:00"},{"alias_kind":"pith_short_12","alias_value":"27VNV4VTAZQJ","created_at":"2026-07-05T10:07:57.754789+00:00"},{"alias_kind":"pith_short_16","alias_value":"27VNV4VTAZQJ4T22","created_at":"2026-07-05T10:07:57.754789+00:00"},{"alias_kind":"pith_short_8","alias_value":"27VNV4VT","created_at":"2026-07-05T10:07:57.754789+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2507.15163","citing_title":"Adaptive Network Security Policies via Belief Aggregation and Rollout","ref_index":36,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/27VNV4VTAZQJ4T224D3JFWZBHL","json":"https://pith.science/pith/27VNV4VTAZQJ4T224D3JFWZBHL.json","graph_json":"https://pith.science/api/pith-number/27VNV4VTAZQJ4T224D3JFWZBHL/graph.json","events_json":"https://pith.science/api/pith-number/27VNV4VTAZQJ4T224D3JFWZBHL/events.json","paper":"https://pith.science/paper/27VNV4VT"},"agent_actions":{"view_html":"https://pith.science/pith/27VNV4VTAZQJ4T224D3JFWZBHL","download_json":"https://pith.science/pith/27VNV4VTAZQJ4T224D3JFWZBHL.json","view_paper":"https://pith.science/paper/27VNV4VT","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2501.19206&json=true","fetch_graph":"https://pith.science/api/pith-number/27VNV4VTAZQJ4T224D3JFWZBHL/graph.json","fetch_events":"https://pith.science/api/pith-number/27VNV4VTAZQJ4T224D3JFWZBHL/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/27VNV4VTAZQJ4T224D3JFWZBHL/action/timestamp_anchor","attest_storage":"https://pith.science/pith/27VNV4VTAZQJ4T224D3JFWZBHL/action/storage_attestation","attest_author":"https://pith.science/pith/27VNV4VTAZQJ4T224D3JFWZBHL/action/author_attestation","sign_citation":"https://pith.science/pith/27VNV4VTAZQJ4T224D3JFWZBHL/action/citation_signature","submit_replication":"https://pith.science/pith/27VNV4VTAZQJ4T224D3JFWZBHL/action/replication_record"}},"created_at":"2026-07-05T10:07:57.754789+00:00","updated_at":"2026-07-05T10:07:57.754789+00:00"}