{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2017:DJIQI4VWSPGE4V3ELJHVNQBAID","short_pith_number":"pith:DJIQI4VW","schema_version":"1.0","canonical_sha256":"1a510472b693cc4e57645a4f56c02040f0f9582f4eec14d8f3bd100b4b5ea51e","source":{"kind":"arxiv","id":"1703.06748","version":4},"attestation_state":"computed","paper":{"title":"Tactics of Adversarial Attack on Deep Reinforcement Learning Agents","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CR","stat.ML"],"primary_cat":"cs.LG","authors_text":"Meng-Li Shih, Ming-Yu Liu, Min Sun, Yen-Chen Lin, Yuan-Hong Liao, Zhang-Wei Hong","submitted_at":"2017-03-08T04:39:34Z","abstract_excerpt":"We introduce two tactics to attack agents trained by deep reinforcement learning algorithms using adversarial examples, namely the strategically-timed attack and the enchanting attack. In the strategically-timed attack, the adversary aims at minimizing the agent's reward by only attacking the agent at a small subset of time steps in an episode. Limiting the attack activity to this subset helps prevent detection of the attack by the agent. We propose a novel method to determine when an adversarial example should be crafted and applied. In the enchanting attack, the adversary aims at luring the "},"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":"1703.06748","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2017-03-08T04:39:34Z","cross_cats_sorted":["cs.CR","stat.ML"],"title_canon_sha256":"cc77a3297e876794f77b16c60b6cb26c6283ba65a7cd9e751c92361c71524ea9","abstract_canon_sha256":"e723a9d4d9787c8baec8c253f16441e13ab06e327d2e893d0e9772592d10fb27"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T00:18:48.190951Z","signature_b64":"9RBvjGiXJeXeArslzH4vOmj2Lj/ZRkmRZCop18tnRp/SxlvHTebU0Yuz6UZmiKUBGgcBO0Ax6euhV9Y68ThoDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"1a510472b693cc4e57645a4f56c02040f0f9582f4eec14d8f3bd100b4b5ea51e","last_reissued_at":"2026-07-05T00:18:48.190470Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T00:18:48.190470Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Tactics of Adversarial Attack on Deep Reinforcement Learning Agents","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CR","stat.ML"],"primary_cat":"cs.LG","authors_text":"Meng-Li Shih, Ming-Yu Liu, Min Sun, Yen-Chen Lin, Yuan-Hong Liao, Zhang-Wei Hong","submitted_at":"2017-03-08T04:39:34Z","abstract_excerpt":"We introduce two tactics to attack agents trained by deep reinforcement learning algorithms using adversarial examples, namely the strategically-timed attack and the enchanting attack. In the strategically-timed attack, the adversary aims at minimizing the agent's reward by only attacking the agent at a small subset of time steps in an episode. Limiting the attack activity to this subset helps prevent detection of the attack by the agent. We propose a novel method to determine when an adversarial example should be crafted and applied. In the enchanting attack, the adversary aims at luring the "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1703.06748","kind":"arxiv","version":4},"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/1703.06748/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":"1703.06748","created_at":"2026-07-05T00:18:48.190532+00:00"},{"alias_kind":"arxiv_version","alias_value":"1703.06748v4","created_at":"2026-07-05T00:18:48.190532+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1703.06748","created_at":"2026-07-05T00:18:48.190532+00:00"},{"alias_kind":"pith_short_12","alias_value":"DJIQI4VWSPGE","created_at":"2026-07-05T00:18:48.190532+00:00"},{"alias_kind":"pith_short_16","alias_value":"DJIQI4VWSPGE4V3E","created_at":"2026-07-05T00:18:48.190532+00:00"},{"alias_kind":"pith_short_8","alias_value":"DJIQI4VW","created_at":"2026-07-05T00:18:48.190532+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":7,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2607.07029","citing_title":"Gimitest: A Comprehensive Tool for Testing Reinforcement Learning Policies","ref_index":16,"is_internal_anchor":true},{"citing_arxiv_id":"2606.09559","citing_title":"Safe-RULE: Safe Reinforcement UnLEarning","ref_index":31,"is_internal_anchor":false},{"citing_arxiv_id":"2606.29867","citing_title":"RoAd-RL: A Unified Library and Benchmark for Robust Adversarial Reinforcement Learning","ref_index":20,"is_internal_anchor":false},{"citing_arxiv_id":"2210.10760","citing_title":"Scaling Laws for Reward Model Overoptimization","ref_index":19,"is_internal_anchor":false},{"citing_arxiv_id":"2604.18874","citing_title":"How Adversarial Environments Mislead Agentic AI?","ref_index":40,"is_internal_anchor":false},{"citing_arxiv_id":"2605.01950","citing_title":"TRAP: Tail-aware Ranking Attack for World-Model Planning","ref_index":32,"is_internal_anchor":false},{"citing_arxiv_id":"2605.02495","citing_title":"Efficient Preference Poisoning Attack on Offline RLHF","ref_index":126,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/DJIQI4VWSPGE4V3ELJHVNQBAID","json":"https://pith.science/pith/DJIQI4VWSPGE4V3ELJHVNQBAID.json","graph_json":"https://pith.science/api/pith-number/DJIQI4VWSPGE4V3ELJHVNQBAID/graph.json","events_json":"https://pith.science/api/pith-number/DJIQI4VWSPGE4V3ELJHVNQBAID/events.json","paper":"https://pith.science/paper/DJIQI4VW"},"agent_actions":{"view_html":"https://pith.science/pith/DJIQI4VWSPGE4V3ELJHVNQBAID","download_json":"https://pith.science/pith/DJIQI4VWSPGE4V3ELJHVNQBAID.json","view_paper":"https://pith.science/paper/DJIQI4VW","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1703.06748&json=true","fetch_graph":"https://pith.science/api/pith-number/DJIQI4VWSPGE4V3ELJHVNQBAID/graph.json","fetch_events":"https://pith.science/api/pith-number/DJIQI4VWSPGE4V3ELJHVNQBAID/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/DJIQI4VWSPGE4V3ELJHVNQBAID/action/timestamp_anchor","attest_storage":"https://pith.science/pith/DJIQI4VWSPGE4V3ELJHVNQBAID/action/storage_attestation","attest_author":"https://pith.science/pith/DJIQI4VWSPGE4V3ELJHVNQBAID/action/author_attestation","sign_citation":"https://pith.science/pith/DJIQI4VWSPGE4V3ELJHVNQBAID/action/citation_signature","submit_replication":"https://pith.science/pith/DJIQI4VWSPGE4V3ELJHVNQBAID/action/replication_record"}},"created_at":"2026-07-05T00:18:48.190532+00:00","updated_at":"2026-07-05T00:18:48.190532+00:00"}