{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:4VGAUT6WYGSUX5YCBGXUI5ZDZD","short_pith_number":"pith:4VGAUT6W","schema_version":"1.0","canonical_sha256":"e54c0a4fd6c1a54bf70209af447723c8d0ca705704637a97f56c6f170d73b51c","source":{"kind":"arxiv","id":"2106.07798","version":1},"attestation_state":"computed","paper":{"title":"Poisoning Deep Reinforcement Learning Agents with In-Distribution Triggers","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CR"],"primary_cat":"cs.LG","authors_text":"Chace Ashcraft, Kiran Karra","submitted_at":"2021-06-14T23:16:41Z","abstract_excerpt":"In this paper, we propose a new data poisoning attack and apply it to deep reinforcement learning agents. Our attack centers on what we call in-distribution triggers, which are triggers native to the data distributions the model will be trained on and deployed in. We outline a simple procedure for embedding these, and other, triggers in deep reinforcement learning agents following a multi-task learning paradigm, and demonstrate in three common reinforcement learning environments. We believe that this work has important implications for the security of deep learning models."},"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":"2106.07798","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2021-06-14T23:16:41Z","cross_cats_sorted":["cs.CR"],"title_canon_sha256":"1ecf86954439b734901d08b9b37cd815124145e0cc0e0aa3428242171c50601e","abstract_canon_sha256":"58b63472a1face20a19437ea3e6ff1a244ca0e2a6d2a21f1345a9e018e0c9c33"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:49:33.801573Z","signature_b64":"zl/F/ibYcB1EUq07syR8IJs5hoHSwZzEyApRt63HW5f2NtNJkAHIcpAEVCoVJme9eVYSdln3uouWhQY9CGrSAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e54c0a4fd6c1a54bf70209af447723c8d0ca705704637a97f56c6f170d73b51c","last_reissued_at":"2026-07-05T02:49:33.801103Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:49:33.801103Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Poisoning Deep Reinforcement Learning Agents with In-Distribution Triggers","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CR"],"primary_cat":"cs.LG","authors_text":"Chace Ashcraft, Kiran Karra","submitted_at":"2021-06-14T23:16:41Z","abstract_excerpt":"In this paper, we propose a new data poisoning attack and apply it to deep reinforcement learning agents. Our attack centers on what we call in-distribution triggers, which are triggers native to the data distributions the model will be trained on and deployed in. We outline a simple procedure for embedding these, and other, triggers in deep reinforcement learning agents following a multi-task learning paradigm, and demonstrate in three common reinforcement learning environments. We believe that this work has important implications for the security of deep learning models."},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2106.07798","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/2106.07798/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":"2106.07798","created_at":"2026-07-05T02:49:33.801157+00:00"},{"alias_kind":"arxiv_version","alias_value":"2106.07798v1","created_at":"2026-07-05T02:49:33.801157+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2106.07798","created_at":"2026-07-05T02:49:33.801157+00:00"},{"alias_kind":"pith_short_12","alias_value":"4VGAUT6WYGSU","created_at":"2026-07-05T02:49:33.801157+00:00"},{"alias_kind":"pith_short_16","alias_value":"4VGAUT6WYGSUX5YC","created_at":"2026-07-05T02:49:33.801157+00:00"},{"alias_kind":"pith_short_8","alias_value":"4VGAUT6W","created_at":"2026-07-05T02:49:33.801157+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.00414","citing_title":"Auditing Near-Optimal Policies Can Be Exponentially Hard: Conditional Query Lower Bounds via Occupancy Rashomon Capacity","ref_index":1,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/4VGAUT6WYGSUX5YCBGXUI5ZDZD","json":"https://pith.science/pith/4VGAUT6WYGSUX5YCBGXUI5ZDZD.json","graph_json":"https://pith.science/api/pith-number/4VGAUT6WYGSUX5YCBGXUI5ZDZD/graph.json","events_json":"https://pith.science/api/pith-number/4VGAUT6WYGSUX5YCBGXUI5ZDZD/events.json","paper":"https://pith.science/paper/4VGAUT6W"},"agent_actions":{"view_html":"https://pith.science/pith/4VGAUT6WYGSUX5YCBGXUI5ZDZD","download_json":"https://pith.science/pith/4VGAUT6WYGSUX5YCBGXUI5ZDZD.json","view_paper":"https://pith.science/paper/4VGAUT6W","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2106.07798&json=true","fetch_graph":"https://pith.science/api/pith-number/4VGAUT6WYGSUX5YCBGXUI5ZDZD/graph.json","fetch_events":"https://pith.science/api/pith-number/4VGAUT6WYGSUX5YCBGXUI5ZDZD/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/4VGAUT6WYGSUX5YCBGXUI5ZDZD/action/timestamp_anchor","attest_storage":"https://pith.science/pith/4VGAUT6WYGSUX5YCBGXUI5ZDZD/action/storage_attestation","attest_author":"https://pith.science/pith/4VGAUT6WYGSUX5YCBGXUI5ZDZD/action/author_attestation","sign_citation":"https://pith.science/pith/4VGAUT6WYGSUX5YCBGXUI5ZDZD/action/citation_signature","submit_replication":"https://pith.science/pith/4VGAUT6WYGSUX5YCBGXUI5ZDZD/action/replication_record"}},"created_at":"2026-07-05T02:49:33.801157+00:00","updated_at":"2026-07-05T02:49:33.801157+00:00"}