{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2019:HJVVGHPKZIBRSR75DJ6I5YZ7MV","short_pith_number":"pith:HJVVGHPK","canonical_record":{"source":{"id":"1907.13548","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-07-31T15:16:00Z","cross_cats_sorted":["cs.CR","stat.ML"],"title_canon_sha256":"b8e7f24029ee150558d0124cc17218a64b9791870101c743620eb9dfea7116f4","abstract_canon_sha256":"eb8b0a14ed0a771ed97c7e137bdc121bbe17920aa8bd2286f8f4f120a4c8fd2b"},"schema_version":"1.0"},"canonical_sha256":"3a6b531deaca031947fd1a7c8ee33f65686e5599ca5b7fb4703b9624645366f6","source":{"kind":"arxiv","id":"1907.13548","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1907.13548","created_at":"2026-07-04T23:50:56Z"},{"alias_kind":"arxiv_version","alias_value":"1907.13548v1","created_at":"2026-07-04T23:50:56Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1907.13548","created_at":"2026-07-04T23:50:56Z"},{"alias_kind":"pith_short_12","alias_value":"HJVVGHPKZIBR","created_at":"2026-07-04T23:50:56Z"},{"alias_kind":"pith_short_16","alias_value":"HJVVGHPKZIBRSR75","created_at":"2026-07-04T23:50:56Z"},{"alias_kind":"pith_short_8","alias_value":"HJVVGHPK","created_at":"2026-07-04T23:50:56Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2019:HJVVGHPKZIBRSR75DJ6I5YZ7MV","target":"record","payload":{"canonical_record":{"source":{"id":"1907.13548","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-07-31T15:16:00Z","cross_cats_sorted":["cs.CR","stat.ML"],"title_canon_sha256":"b8e7f24029ee150558d0124cc17218a64b9791870101c743620eb9dfea7116f4","abstract_canon_sha256":"eb8b0a14ed0a771ed97c7e137bdc121bbe17920aa8bd2286f8f4f120a4c8fd2b"},"schema_version":"1.0"},"canonical_sha256":"3a6b531deaca031947fd1a7c8ee33f65686e5599ca5b7fb4703b9624645366f6","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-04T23:50:56.338483Z","signature_b64":"3KOXJkmZAP53nSA+TZMp51tKszKJ7m+d42JuqQZuFu5fE6TEmmRdOzN7QpsK7QtUdUfcotszaITeHYlC0dBqDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3a6b531deaca031947fd1a7c8ee33f65686e5599ca5b7fb4703b9624645366f6","last_reissued_at":"2026-07-04T23:50:56.337988Z","signature_status":"signed_v1","first_computed_at":"2026-07-04T23:50:56.337988Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"1907.13548","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":"2026-07-04T23:50:56Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"CMqHUiHCq0GyLPrNdXgHIZg33ympcDbvAaJlZl59yd/GtlR3yH4HMPS5JMWdAb0Ukhq/rumyhtHHK2u6tth0Cg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-04T19:28:53.986725Z"},"content_sha256":"01ff4a8ccac0a8af6b643581de9b9e8ef402da68edb004e7744437656c8aaa82","schema_version":"1.0","event_id":"sha256:01ff4a8ccac0a8af6b643581de9b9e8ef402da68edb004e7744437656c8aaa82"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2019:HJVVGHPKZIBRSR75DJ6I5YZ7MV","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Optimal Attacks on Reinforcement Learning Policies","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CR","stat.ML"],"primary_cat":"cs.LG","authors_text":"Alessio Russo, Alexandre Proutiere","submitted_at":"2019-07-31T15:16:00Z","abstract_excerpt":"Control policies, trained using the Deep Reinforcement Learning, have been recently shown to be vulnerable to adversarial attacks introducing even very small perturbations to the policy input. The attacks proposed so far have been designed using heuristics, and build on existing adversarial example crafting techniques used to dupe classifiers in supervised learning. In contrast, this paper investigates the problem of devising optimal attacks, depending on a well-defined attacker's objective, e.g., to minimize the main agent average reward. When the policy and the system dynamics, as well as re"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1907.13548","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/1907.13548/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-04T23:50:56Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"jZsJ4DXYOhthd6Q/NJwN9TjUOs6iDBBDuxS02Lua6qKM+ZOcMLn5s1ua5tfUC4SwAPVYc3EMIBQ1zdURSVEVDw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-04T19:28:53.987213Z"},"content_sha256":"fcde81b06299e5667d6eef04796b09754f5d20fbde8e34d59d14fb3d89ef2be9","schema_version":"1.0","event_id":"sha256:fcde81b06299e5667d6eef04796b09754f5d20fbde8e34d59d14fb3d89ef2be9"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/HJVVGHPKZIBRSR75DJ6I5YZ7MV/bundle.json","state_url":"https://pith.science/pith/HJVVGHPKZIBRSR75DJ6I5YZ7MV/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/HJVVGHPKZIBRSR75DJ6I5YZ7MV/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":"2026-08-04T19:28:53Z","links":{"resolver":"https://pith.science/pith/HJVVGHPKZIBRSR75DJ6I5YZ7MV","bundle":"https://pith.science/pith/HJVVGHPKZIBRSR75DJ6I5YZ7MV/bundle.json","state":"https://pith.science/pith/HJVVGHPKZIBRSR75DJ6I5YZ7MV/state.json","well_known_bundle":"https://pith.science/.well-known/pith/HJVVGHPKZIBRSR75DJ6I5YZ7MV/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2019:HJVVGHPKZIBRSR75DJ6I5YZ7MV","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":"eb8b0a14ed0a771ed97c7e137bdc121bbe17920aa8bd2286f8f4f120a4c8fd2b","cross_cats_sorted":["cs.CR","stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-07-31T15:16:00Z","title_canon_sha256":"b8e7f24029ee150558d0124cc17218a64b9791870101c743620eb9dfea7116f4"},"schema_version":"1.0","source":{"id":"1907.13548","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1907.13548","created_at":"2026-07-04T23:50:56Z"},{"alias_kind":"arxiv_version","alias_value":"1907.13548v1","created_at":"2026-07-04T23:50:56Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1907.13548","created_at":"2026-07-04T23:50:56Z"},{"alias_kind":"pith_short_12","alias_value":"HJVVGHPKZIBR","created_at":"2026-07-04T23:50:56Z"},{"alias_kind":"pith_short_16","alias_value":"HJVVGHPKZIBRSR75","created_at":"2026-07-04T23:50:56Z"},{"alias_kind":"pith_short_8","alias_value":"HJVVGHPK","created_at":"2026-07-04T23:50:56Z"}],"graph_snapshots":[{"event_id":"sha256:fcde81b06299e5667d6eef04796b09754f5d20fbde8e34d59d14fb3d89ef2be9","target":"graph","created_at":"2026-07-04T23:50:56Z","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"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/1907.13548/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Control policies, trained using the Deep Reinforcement Learning, have been recently shown to be vulnerable to adversarial attacks introducing even very small perturbations to the policy input. The attacks proposed so far have been designed using heuristics, and build on existing adversarial example crafting techniques used to dupe classifiers in supervised learning. In contrast, this paper investigates the problem of devising optimal attacks, depending on a well-defined attacker's objective, e.g., to minimize the main agent average reward. When the policy and the system dynamics, as well as re","authors_text":"Alessio Russo, Alexandre Proutiere","cross_cats":["cs.CR","stat.ML"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-07-31T15:16:00Z","title":"Optimal Attacks on Reinforcement Learning Policies"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1907.13548","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":"sha256:01ff4a8ccac0a8af6b643581de9b9e8ef402da68edb004e7744437656c8aaa82","target":"record","created_at":"2026-07-04T23:50:56Z","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":"eb8b0a14ed0a771ed97c7e137bdc121bbe17920aa8bd2286f8f4f120a4c8fd2b","cross_cats_sorted":["cs.CR","stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-07-31T15:16:00Z","title_canon_sha256":"b8e7f24029ee150558d0124cc17218a64b9791870101c743620eb9dfea7116f4"},"schema_version":"1.0","source":{"id":"1907.13548","kind":"arxiv","version":1}},"canonical_sha256":"3a6b531deaca031947fd1a7c8ee33f65686e5599ca5b7fb4703b9624645366f6","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"3a6b531deaca031947fd1a7c8ee33f65686e5599ca5b7fb4703b9624645366f6","first_computed_at":"2026-07-04T23:50:56.337988Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-04T23:50:56.337988Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"3KOXJkmZAP53nSA+TZMp51tKszKJ7m+d42JuqQZuFu5fE6TEmmRdOzN7QpsK7QtUdUfcotszaITeHYlC0dBqDQ==","signature_status":"signed_v1","signed_at":"2026-07-04T23:50:56.338483Z","signed_message":"canonical_sha256_bytes"},"source_id":"1907.13548","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:01ff4a8ccac0a8af6b643581de9b9e8ef402da68edb004e7744437656c8aaa82","sha256:fcde81b06299e5667d6eef04796b09754f5d20fbde8e34d59d14fb3d89ef2be9"],"state_sha256":"09e53bdd238f93a777a3c038623f76a3c3bbeb527923a4d9c5230e56a43cdf02"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"lNIQIqL4Ts0tvex3FFnJlE3YfsUeLWjO7I5ndSNmPHdwpmPjNFv5ouhDzkT87W2C3HMyC2H9Gp60fkAiLFx6BA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-04T19:28:53.990917Z","bundle_sha256":"bdd73ca6768f7388f42271ce27bb01272ed1603ece0cf8074ab6f783ccf8caf5"}}