{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2017:7BYQSF6CDZ4XP7N37MRTELWYIV","short_pith_number":"pith:7BYQSF6C","canonical_record":{"source":{"id":"1712.09344","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2017-12-23T23:57:55Z","cross_cats_sorted":[],"title_canon_sha256":"83d42e82814a75e5ddcc1261aada02702a9b3be9d1e61e6c2b8a13aeebd15dae","abstract_canon_sha256":"c7b9b4b945383471120f1d154b0defcf9cb956955c99a2bfc0e52b556141ad52"},"schema_version":"1.0"},"canonical_sha256":"f8710917c21e7977fdbbfb23322ed84544e73ddecca8378d995070f7c89ee9eb","source":{"kind":"arxiv","id":"1712.09344","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1712.09344","created_at":"2026-05-18T00:27:08Z"},{"alias_kind":"arxiv_version","alias_value":"1712.09344v1","created_at":"2026-05-18T00:27:08Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1712.09344","created_at":"2026-05-18T00:27:08Z"},{"alias_kind":"pith_short_12","alias_value":"7BYQSF6CDZ4X","created_at":"2026-05-18T12:31:03Z"},{"alias_kind":"pith_short_16","alias_value":"7BYQSF6CDZ4XP7N3","created_at":"2026-05-18T12:31:03Z"},{"alias_kind":"pith_short_8","alias_value":"7BYQSF6C","created_at":"2026-05-18T12:31:03Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2017:7BYQSF6CDZ4XP7N37MRTELWYIV","target":"record","payload":{"canonical_record":{"source":{"id":"1712.09344","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2017-12-23T23:57:55Z","cross_cats_sorted":[],"title_canon_sha256":"83d42e82814a75e5ddcc1261aada02702a9b3be9d1e61e6c2b8a13aeebd15dae","abstract_canon_sha256":"c7b9b4b945383471120f1d154b0defcf9cb956955c99a2bfc0e52b556141ad52"},"schema_version":"1.0"},"canonical_sha256":"f8710917c21e7977fdbbfb23322ed84544e73ddecca8378d995070f7c89ee9eb","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T00:27:08.714755Z","signature_b64":"GDXep+e3LhDJqtBFBU/OTwkg8zkCfsMJ1s4gvfJeN+nwobVYl1HpeJtDXMPjpd2gv8fXaX0/x+Scji1Ccr4jCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f8710917c21e7977fdbbfb23322ed84544e73ddecca8378d995070f7c89ee9eb","last_reissued_at":"2026-05-18T00:27:08.714182Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T00:27:08.714182Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"1712.09344","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-18T00:27:08Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"sTP5e+y7gl1TXQ/AM96VsfV3HJhLRIa9l1Erh0mp1QJWjyIlcMPRiyeyPeoeIoLsEF07b1Cn7LhcmsxdzYcADg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-25T21:02:33.818751Z"},"content_sha256":"863a0b1cad2a1a985129ef60c2b1a57a71f339585dfd874603860b60dbf8d044","schema_version":"1.0","event_id":"sha256:863a0b1cad2a1a985129ef60c2b1a57a71f339585dfd874603860b60dbf8d044"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2017:7BYQSF6CDZ4XP7N37MRTELWYIV","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Whatever Does Not Kill Deep Reinforcement Learning, Makes It Stronger","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Arslan Munir, Vahid Behzadan","submitted_at":"2017-12-23T23:57:55Z","abstract_excerpt":"Recent developments have established the vulnerability of deep Reinforcement Learning (RL) to policy manipulation attacks via adversarial perturbations. In this paper, we investigate the robustness and resilience of deep RL to training-time and test-time attacks. Through experimental results, we demonstrate that under noncontiguous training-time attacks, Deep Q-Network (DQN) agents can recover and adapt to the adversarial conditions by reactively adjusting the policy. Our results also show that policies learned under adversarial perturbations are more robust to test-time attacks. Furthermore, "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1712.09344","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":""},"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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-18T00:27:08Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"v02Tmocm9ahljS7IO1pWgUbNMw7B3dSEzkVzc9givgM1aSykjAr82/X+XBDmni4O3ApEHeU7JVSBOo5Sah+3Bg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-25T21:02:33.819463Z"},"content_sha256":"ad1195887a8db90f094af4399c7c8458f484ca27adb9109d58a78a6cb6cd501a","schema_version":"1.0","event_id":"sha256:ad1195887a8db90f094af4399c7c8458f484ca27adb9109d58a78a6cb6cd501a"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/7BYQSF6CDZ4XP7N37MRTELWYIV/bundle.json","state_url":"https://pith.science/pith/7BYQSF6CDZ4XP7N37MRTELWYIV/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/7BYQSF6CDZ4XP7N37MRTELWYIV/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-05-25T21:02:33Z","links":{"resolver":"https://pith.science/pith/7BYQSF6CDZ4XP7N37MRTELWYIV","bundle":"https://pith.science/pith/7BYQSF6CDZ4XP7N37MRTELWYIV/bundle.json","state":"https://pith.science/pith/7BYQSF6CDZ4XP7N37MRTELWYIV/state.json","well_known_bundle":"https://pith.science/.well-known/pith/7BYQSF6CDZ4XP7N37MRTELWYIV/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2017:7BYQSF6CDZ4XP7N37MRTELWYIV","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"c7b9b4b945383471120f1d154b0defcf9cb956955c99a2bfc0e52b556141ad52","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2017-12-23T23:57:55Z","title_canon_sha256":"83d42e82814a75e5ddcc1261aada02702a9b3be9d1e61e6c2b8a13aeebd15dae"},"schema_version":"1.0","source":{"id":"1712.09344","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1712.09344","created_at":"2026-05-18T00:27:08Z"},{"alias_kind":"arxiv_version","alias_value":"1712.09344v1","created_at":"2026-05-18T00:27:08Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1712.09344","created_at":"2026-05-18T00:27:08Z"},{"alias_kind":"pith_short_12","alias_value":"7BYQSF6CDZ4X","created_at":"2026-05-18T12:31:03Z"},{"alias_kind":"pith_short_16","alias_value":"7BYQSF6CDZ4XP7N3","created_at":"2026-05-18T12:31:03Z"},{"alias_kind":"pith_short_8","alias_value":"7BYQSF6C","created_at":"2026-05-18T12:31:03Z"}],"graph_snapshots":[{"event_id":"sha256:ad1195887a8db90f094af4399c7c8458f484ca27adb9109d58a78a6cb6cd501a","target":"graph","created_at":"2026-05-18T00:27:08Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"paper":{"abstract_excerpt":"Recent developments have established the vulnerability of deep Reinforcement Learning (RL) to policy manipulation attacks via adversarial perturbations. In this paper, we investigate the robustness and resilience of deep RL to training-time and test-time attacks. Through experimental results, we demonstrate that under noncontiguous training-time attacks, Deep Q-Network (DQN) agents can recover and adapt to the adversarial conditions by reactively adjusting the policy. Our results also show that policies learned under adversarial perturbations are more robust to test-time attacks. Furthermore, ","authors_text":"Arslan Munir, Vahid Behzadan","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2017-12-23T23:57:55Z","title":"Whatever Does Not Kill Deep Reinforcement Learning, Makes It Stronger"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1712.09344","kind":"arxiv","version":1},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:863a0b1cad2a1a985129ef60c2b1a57a71f339585dfd874603860b60dbf8d044","target":"record","created_at":"2026-05-18T00:27:08Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"c7b9b4b945383471120f1d154b0defcf9cb956955c99a2bfc0e52b556141ad52","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2017-12-23T23:57:55Z","title_canon_sha256":"83d42e82814a75e5ddcc1261aada02702a9b3be9d1e61e6c2b8a13aeebd15dae"},"schema_version":"1.0","source":{"id":"1712.09344","kind":"arxiv","version":1}},"canonical_sha256":"f8710917c21e7977fdbbfb23322ed84544e73ddecca8378d995070f7c89ee9eb","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"f8710917c21e7977fdbbfb23322ed84544e73ddecca8378d995070f7c89ee9eb","first_computed_at":"2026-05-18T00:27:08.714182Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-18T00:27:08.714182Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"GDXep+e3LhDJqtBFBU/OTwkg8zkCfsMJ1s4gvfJeN+nwobVYl1HpeJtDXMPjpd2gv8fXaX0/x+Scji1Ccr4jCg==","signature_status":"signed_v1","signed_at":"2026-05-18T00:27:08.714755Z","signed_message":"canonical_sha256_bytes"},"source_id":"1712.09344","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:863a0b1cad2a1a985129ef60c2b1a57a71f339585dfd874603860b60dbf8d044","sha256:ad1195887a8db90f094af4399c7c8458f484ca27adb9109d58a78a6cb6cd501a"],"state_sha256":"4027077803834a0ac79c0960e33678f38755f4ef6dbafa27b14d71159e676cf5"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"BEZVKdJlzynUclP8KSS1zpjd+1VMpt3SSMBpBax2YYMG451fslcwm/MDdvi2NvNX7jU0C0I8GOhwSF/DAufHDg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-25T21:02:33.823307Z","bundle_sha256":"9e776cd5b56dc04dedc6738f2c4e926aaabef3926464188970d45f92e3a19bcd"}}