{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:XEU2Y2Q5W7MEQOTKV2MH2AFZFX","short_pith_number":"pith:XEU2Y2Q5","schema_version":"1.0","canonical_sha256":"b929ac6a1db7d8483a6aae987d00b92df778abadf16aa7abb777fae235dee851","source":{"kind":"arxiv","id":"2401.02652","version":1},"attestation_state":"computed","paper":{"title":"Adaptive Discounting of Training Time Attacks","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.AI","cs.CR"],"primary_cat":"cs.LG","authors_text":"Abhay Aradhya, Chai Quek, Ridhima Bector, Zinovi Rabinovich","submitted_at":"2024-01-05T06:03:14Z","abstract_excerpt":"Among the most insidious attacks on Reinforcement Learning (RL) solutions are training-time attacks (TTAs) that create loopholes and backdoors in the learned behaviour. Not limited to a simple disruption, constructive TTAs (C-TTAs) are now available, where the attacker forces a specific, target behaviour upon a training RL agent (victim). However, even state-of-the-art C-TTAs focus on target behaviours that could be naturally adopted by the victim if not for a particular feature of the environment dynamics, which C-TTAs exploit. In this work, we show that a C-TTA is possible even when the targ"},"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":"2401.02652","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.LG","submitted_at":"2024-01-05T06:03:14Z","cross_cats_sorted":["cs.AI","cs.CR"],"title_canon_sha256":"bd7b9a41a3546f935c719243e07dd5157b9064abe8a3129f79768425830a6b8a","abstract_canon_sha256":"c1391969c54cdfc31ea652bee10fd65e0c0d29aee4ea5cf03274ec99142bc48c"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:30:29.121468Z","signature_b64":"pAyV02O+P/YtTLXA+o40+S4BdTcpaW9qumW+MDYy38h5SBbHKGSClfIphu1qeKB1Qu8cZVfUN/lsEIUNRag4DQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b929ac6a1db7d8483a6aae987d00b92df778abadf16aa7abb777fae235dee851","last_reissued_at":"2026-07-05T07:30:29.120983Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:30:29.120983Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Adaptive Discounting of Training Time Attacks","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.AI","cs.CR"],"primary_cat":"cs.LG","authors_text":"Abhay Aradhya, Chai Quek, Ridhima Bector, Zinovi Rabinovich","submitted_at":"2024-01-05T06:03:14Z","abstract_excerpt":"Among the most insidious attacks on Reinforcement Learning (RL) solutions are training-time attacks (TTAs) that create loopholes and backdoors in the learned behaviour. Not limited to a simple disruption, constructive TTAs (C-TTAs) are now available, where the attacker forces a specific, target behaviour upon a training RL agent (victim). However, even state-of-the-art C-TTAs focus on target behaviours that could be naturally adopted by the victim if not for a particular feature of the environment dynamics, which C-TTAs exploit. In this work, we show that a C-TTA is possible even when the targ"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2401.02652","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/2401.02652/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":"2401.02652","created_at":"2026-07-05T07:30:29.121043+00:00"},{"alias_kind":"arxiv_version","alias_value":"2401.02652v1","created_at":"2026-07-05T07:30:29.121043+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2401.02652","created_at":"2026-07-05T07:30:29.121043+00:00"},{"alias_kind":"pith_short_12","alias_value":"XEU2Y2Q5W7ME","created_at":"2026-07-05T07:30:29.121043+00:00"},{"alias_kind":"pith_short_16","alias_value":"XEU2Y2Q5W7MEQOTK","created_at":"2026-07-05T07:30:29.121043+00:00"},{"alias_kind":"pith_short_8","alias_value":"XEU2Y2Q5","created_at":"2026-07-05T07:30:29.121043+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/XEU2Y2Q5W7MEQOTKV2MH2AFZFX","json":"https://pith.science/pith/XEU2Y2Q5W7MEQOTKV2MH2AFZFX.json","graph_json":"https://pith.science/api/pith-number/XEU2Y2Q5W7MEQOTKV2MH2AFZFX/graph.json","events_json":"https://pith.science/api/pith-number/XEU2Y2Q5W7MEQOTKV2MH2AFZFX/events.json","paper":"https://pith.science/paper/XEU2Y2Q5"},"agent_actions":{"view_html":"https://pith.science/pith/XEU2Y2Q5W7MEQOTKV2MH2AFZFX","download_json":"https://pith.science/pith/XEU2Y2Q5W7MEQOTKV2MH2AFZFX.json","view_paper":"https://pith.science/paper/XEU2Y2Q5","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2401.02652&json=true","fetch_graph":"https://pith.science/api/pith-number/XEU2Y2Q5W7MEQOTKV2MH2AFZFX/graph.json","fetch_events":"https://pith.science/api/pith-number/XEU2Y2Q5W7MEQOTKV2MH2AFZFX/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/XEU2Y2Q5W7MEQOTKV2MH2AFZFX/action/timestamp_anchor","attest_storage":"https://pith.science/pith/XEU2Y2Q5W7MEQOTKV2MH2AFZFX/action/storage_attestation","attest_author":"https://pith.science/pith/XEU2Y2Q5W7MEQOTKV2MH2AFZFX/action/author_attestation","sign_citation":"https://pith.science/pith/XEU2Y2Q5W7MEQOTKV2MH2AFZFX/action/citation_signature","submit_replication":"https://pith.science/pith/XEU2Y2Q5W7MEQOTKV2MH2AFZFX/action/replication_record"}},"created_at":"2026-07-05T07:30:29.121043+00:00","updated_at":"2026-07-05T07:30:29.121043+00:00"}