{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2019:XLTD2EFNCXAMDAEX3AERY7PBOI","short_pith_number":"pith:XLTD2EFN","canonical_record":{"source":{"id":"1911.00887","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-11-03T13:44:42Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"af42d0e37e0a460bb0c9f520d96b3b6c8badc27c120dc90f1b9736f765cf767f","abstract_canon_sha256":"524da01f9e3c2ebd8ceb483dc3eb0c519e6201c3b15e2e76272f96ab8ecbacc5"},"schema_version":"1.0"},"canonical_sha256":"bae63d10ad15c0c18097d8091c7de172323e9ced6810f201fceacd9fe87ff144","source":{"kind":"arxiv","id":"1911.00887","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1911.00887","created_at":"2026-07-05T00:21:15Z"},{"alias_kind":"arxiv_version","alias_value":"1911.00887v3","created_at":"2026-07-05T00:21:15Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1911.00887","created_at":"2026-07-05T00:21:15Z"},{"alias_kind":"pith_short_12","alias_value":"XLTD2EFNCXAM","created_at":"2026-07-05T00:21:15Z"},{"alias_kind":"pith_short_16","alias_value":"XLTD2EFNCXAMDAEX","created_at":"2026-07-05T00:21:15Z"},{"alias_kind":"pith_short_8","alias_value":"XLTD2EFN","created_at":"2026-07-05T00:21:15Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2019:XLTD2EFNCXAMDAEX3AERY7PBOI","target":"record","payload":{"canonical_record":{"source":{"id":"1911.00887","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-11-03T13:44:42Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"af42d0e37e0a460bb0c9f520d96b3b6c8badc27c120dc90f1b9736f765cf767f","abstract_canon_sha256":"524da01f9e3c2ebd8ceb483dc3eb0c519e6201c3b15e2e76272f96ab8ecbacc5"},"schema_version":"1.0"},"canonical_sha256":"bae63d10ad15c0c18097d8091c7de172323e9ced6810f201fceacd9fe87ff144","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T00:21:15.856439Z","signature_b64":"mppvt1OOGSXo1XE379GS/3tRvjQuZSmPxHUaMxB1HuZSEh2QDzBBsm8XuahYBnVqN8BRC7VauVbqnKhUjje9BA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"bae63d10ad15c0c18097d8091c7de172323e9ced6810f201fceacd9fe87ff144","last_reissued_at":"2026-07-05T00:21:15.855892Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T00:21:15.855892Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"1911.00887","source_version":3,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T00:21:15Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"bjEW0RC3yJe/31MSDvnxKaz+QKjke9K0QU5ONayhNP2Rvl9ndcxDGkc0efl7Lq2MRSk7OEPHTyGmQJGp3HxADA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T21:14:28.642653Z"},"content_sha256":"2280042b75126f6222bf6243d344bc83d074dd19996b58a09b955c8b8f0b8eed","schema_version":"1.0","event_id":"sha256:2280042b75126f6222bf6243d344bc83d074dd19996b58a09b955c8b8f0b8eed"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2019:XLTD2EFNCXAMDAEX3AERY7PBOI","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Online Robustness Training for Deep Reinforcement Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Marc Fischer, Martin Vechev, Matthew Mirman, Steven Stalder","submitted_at":"2019-11-03T13:44:42Z","abstract_excerpt":"In deep reinforcement learning (RL), adversarial attacks can trick an agent into unwanted states and disrupt training. We propose a system called Robust Student-DQN (RS-DQN), which permits online robustness training alongside Q networks, while preserving competitive performance. We show that RS-DQN can be combined with (i) state-of-the-art adversarial training and (ii) provably robust training to obtain an agent that is resilient to strong attacks during training and evaluation."},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1911.00887","kind":"arxiv","version":3},"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/1911.00887/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":"2026-07-05T00:21:15Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"98Pxgh+f8kMl2jY4Er3pg6O8s7fbF42oXhRt21ICDYwfDMoC0zF+imCbjOysbponBEUzyLRMIMUP6YSP0utyCw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T21:14:28.643195Z"},"content_sha256":"13569ccd2a79288eac2fa834a4f4e035e84b5a259018007be3dff83bb9d0e508","schema_version":"1.0","event_id":"sha256:13569ccd2a79288eac2fa834a4f4e035e84b5a259018007be3dff83bb9d0e508"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/XLTD2EFNCXAMDAEX3AERY7PBOI/bundle.json","state_url":"https://pith.science/pith/XLTD2EFNCXAMDAEX3AERY7PBOI/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/XLTD2EFNCXAMDAEX3AERY7PBOI/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":"2026-08-08T21:14:28Z","links":{"resolver":"https://pith.science/pith/XLTD2EFNCXAMDAEX3AERY7PBOI","bundle":"https://pith.science/pith/XLTD2EFNCXAMDAEX3AERY7PBOI/bundle.json","state":"https://pith.science/pith/XLTD2EFNCXAMDAEX3AERY7PBOI/state.json","well_known_bundle":"https://pith.science/.well-known/pith/XLTD2EFNCXAMDAEX3AERY7PBOI/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2019:XLTD2EFNCXAMDAEX3AERY7PBOI","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":"524da01f9e3c2ebd8ceb483dc3eb0c519e6201c3b15e2e76272f96ab8ecbacc5","cross_cats_sorted":["stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-11-03T13:44:42Z","title_canon_sha256":"af42d0e37e0a460bb0c9f520d96b3b6c8badc27c120dc90f1b9736f765cf767f"},"schema_version":"1.0","source":{"id":"1911.00887","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1911.00887","created_at":"2026-07-05T00:21:15Z"},{"alias_kind":"arxiv_version","alias_value":"1911.00887v3","created_at":"2026-07-05T00:21:15Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1911.00887","created_at":"2026-07-05T00:21:15Z"},{"alias_kind":"pith_short_12","alias_value":"XLTD2EFNCXAM","created_at":"2026-07-05T00:21:15Z"},{"alias_kind":"pith_short_16","alias_value":"XLTD2EFNCXAMDAEX","created_at":"2026-07-05T00:21:15Z"},{"alias_kind":"pith_short_8","alias_value":"XLTD2EFN","created_at":"2026-07-05T00:21:15Z"}],"graph_snapshots":[{"event_id":"sha256:13569ccd2a79288eac2fa834a4f4e035e84b5a259018007be3dff83bb9d0e508","target":"graph","created_at":"2026-07-05T00:21:15Z","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/1911.00887/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"In deep reinforcement learning (RL), adversarial attacks can trick an agent into unwanted states and disrupt training. We propose a system called Robust Student-DQN (RS-DQN), which permits online robustness training alongside Q networks, while preserving competitive performance. We show that RS-DQN can be combined with (i) state-of-the-art adversarial training and (ii) provably robust training to obtain an agent that is resilient to strong attacks during training and evaluation.","authors_text":"Marc Fischer, Martin Vechev, Matthew Mirman, Steven Stalder","cross_cats":["stat.ML"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-11-03T13:44:42Z","title":"Online Robustness Training for Deep Reinforcement Learning"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1911.00887","kind":"arxiv","version":3},"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":"sha256:2280042b75126f6222bf6243d344bc83d074dd19996b58a09b955c8b8f0b8eed","target":"record","created_at":"2026-07-05T00:21:15Z","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":"524da01f9e3c2ebd8ceb483dc3eb0c519e6201c3b15e2e76272f96ab8ecbacc5","cross_cats_sorted":["stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-11-03T13:44:42Z","title_canon_sha256":"af42d0e37e0a460bb0c9f520d96b3b6c8badc27c120dc90f1b9736f765cf767f"},"schema_version":"1.0","source":{"id":"1911.00887","kind":"arxiv","version":3}},"canonical_sha256":"bae63d10ad15c0c18097d8091c7de172323e9ced6810f201fceacd9fe87ff144","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"bae63d10ad15c0c18097d8091c7de172323e9ced6810f201fceacd9fe87ff144","first_computed_at":"2026-07-05T00:21:15.855892Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T00:21:15.855892Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"mppvt1OOGSXo1XE379GS/3tRvjQuZSmPxHUaMxB1HuZSEh2QDzBBsm8XuahYBnVqN8BRC7VauVbqnKhUjje9BA==","signature_status":"signed_v1","signed_at":"2026-07-05T00:21:15.856439Z","signed_message":"canonical_sha256_bytes"},"source_id":"1911.00887","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:2280042b75126f6222bf6243d344bc83d074dd19996b58a09b955c8b8f0b8eed","sha256:13569ccd2a79288eac2fa834a4f4e035e84b5a259018007be3dff83bb9d0e508"],"state_sha256":"433f87e1ce91fb8d9e271be239226b41a268dd8fddb5b405a87221446e4e2954"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"dtzm2499yRginVZqWcuArWYLnuH5sZJ/j6ZkR8VwiiRitTnBXI2JXz2YGcXT4tvFU7jaI62b1CPbczH+Mqs4CQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-08T21:14:28.648016Z","bundle_sha256":"ad1a3624c682e522989886c65c39f18a0ebc2967d37c176fb759e7bececb79bf"}}