{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2023:VWUMGNQ5GKFDN3IFWT5BFM5CMV","short_pith_number":"pith:VWUMGNQ5","canonical_record":{"source":{"id":"2309.03791","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2023-09-07T15:41:45Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"34b9da94183514ae4171c9d79b88ee9e83b2bb15634bf1c68b5b4960a09a84f2","abstract_canon_sha256":"1b867e5c899950286e67e34ecef2388f831e01c52765ae15f75072fb7c952871"},"schema_version":"1.0"},"canonical_sha256":"ada8c3361d328a36ed05b4fa12b3a2656832ca0d4d2db17f672d004f1209f366","source":{"kind":"arxiv","id":"2309.03791","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2309.03791","created_at":"2026-07-05T11:42:19Z"},{"alias_kind":"arxiv_version","alias_value":"2309.03791v3","created_at":"2026-07-05T11:42:19Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2309.03791","created_at":"2026-07-05T11:42:19Z"},{"alias_kind":"pith_short_12","alias_value":"VWUMGNQ5GKFD","created_at":"2026-07-05T11:42:19Z"},{"alias_kind":"pith_short_16","alias_value":"VWUMGNQ5GKFDN3IF","created_at":"2026-07-05T11:42:19Z"},{"alias_kind":"pith_short_8","alias_value":"VWUMGNQ5","created_at":"2026-07-05T11:42:19Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2023:VWUMGNQ5GKFDN3IFWT5BFM5CMV","target":"record","payload":{"canonical_record":{"source":{"id":"2309.03791","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2023-09-07T15:41:45Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"34b9da94183514ae4171c9d79b88ee9e83b2bb15634bf1c68b5b4960a09a84f2","abstract_canon_sha256":"1b867e5c899950286e67e34ecef2388f831e01c52765ae15f75072fb7c952871"},"schema_version":"1.0"},"canonical_sha256":"ada8c3361d328a36ed05b4fa12b3a2656832ca0d4d2db17f672d004f1209f366","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:42:19.812644Z","signature_b64":"VHbUiL3kLq7Z4meyeyjkl7Das2Qibl1SbyN+0hk9DFmYfNveN+M4STwwtpQYn8SeCOe6f0t8TY09ZNe5alU2Aw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ada8c3361d328a36ed05b4fa12b3a2656832ca0d4d2db17f672d004f1209f366","last_reissued_at":"2026-07-05T11:42:19.812231Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:42:19.812231Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2309.03791","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":"2026-07-05T11:42:19Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"+SDrtTRPjP/EEpaiIWoeCN6fNdb3qSLDSmgxXvku0HRnHmj17+UpMH5MnbDgW7G8lhG6i4BfiDqFvku+hE5FCg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-09T04:36:46.870818Z"},"content_sha256":"245993df9ef5a7778afd45c9bd5b51b1eb934b731ea59611d357423f7fdf1cd2","schema_version":"1.0","event_id":"sha256:245993df9ef5a7778afd45c9bd5b51b1eb934b731ea59611d357423f7fdf1cd2"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2023:VWUMGNQ5GKFDN3IFWT5BFM5CMV","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Optimal Transport Regularized Divergences: Application to Adversarial Robustness","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Jeremiah Birrell, Reza Ebrahimi","submitted_at":"2023-09-07T15:41:45Z","abstract_excerpt":"We introduce a new class of optimal-transport-regularized divergences, $D^c$, constructed via an infimal convolution between an information divergence, $D$, and an optimal-transport (OT) cost, $C$, and study their use in distributionally robust optimization (DRO). In particular, we propose the $ARMOR_D$ methods as novel approaches to enhancing the adversarial robustness of deep learning models. These DRO-based methods are defined by minimizing the maximum expected loss over a $D^c$-neighborhood of the empirical distribution of the training data. Viewed as a tool for constructing adversarial sa"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2309.03791","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/2309.03791/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":"2026-07-05T11:42:19Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"1zQpJQ8iVLEeaVaJq+aBNuvGoloxHngF7gtFDZ9GvF1vqRjhykI6tA3PBqGKfc7ZyrIKUlqTYwlPIqHf6TXAAg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-09T04:36:46.871303Z"},"content_sha256":"e9c4632167dc975a315dc6e19e409d3d5d33ade389bd3574fd4ec192993e897f","schema_version":"1.0","event_id":"sha256:e9c4632167dc975a315dc6e19e409d3d5d33ade389bd3574fd4ec192993e897f"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/VWUMGNQ5GKFDN3IFWT5BFM5CMV/bundle.json","state_url":"https://pith.science/pith/VWUMGNQ5GKFDN3IFWT5BFM5CMV/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/VWUMGNQ5GKFDN3IFWT5BFM5CMV/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":"2026-08-09T04:36:46Z","links":{"resolver":"https://pith.science/pith/VWUMGNQ5GKFDN3IFWT5BFM5CMV","bundle":"https://pith.science/pith/VWUMGNQ5GKFDN3IFWT5BFM5CMV/bundle.json","state":"https://pith.science/pith/VWUMGNQ5GKFDN3IFWT5BFM5CMV/state.json","well_known_bundle":"https://pith.science/.well-known/pith/VWUMGNQ5GKFDN3IFWT5BFM5CMV/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:VWUMGNQ5GKFDN3IFWT5BFM5CMV","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":"1b867e5c899950286e67e34ecef2388f831e01c52765ae15f75072fb7c952871","cross_cats_sorted":["stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2023-09-07T15:41:45Z","title_canon_sha256":"34b9da94183514ae4171c9d79b88ee9e83b2bb15634bf1c68b5b4960a09a84f2"},"schema_version":"1.0","source":{"id":"2309.03791","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2309.03791","created_at":"2026-07-05T11:42:19Z"},{"alias_kind":"arxiv_version","alias_value":"2309.03791v3","created_at":"2026-07-05T11:42:19Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2309.03791","created_at":"2026-07-05T11:42:19Z"},{"alias_kind":"pith_short_12","alias_value":"VWUMGNQ5GKFD","created_at":"2026-07-05T11:42:19Z"},{"alias_kind":"pith_short_16","alias_value":"VWUMGNQ5GKFDN3IF","created_at":"2026-07-05T11:42:19Z"},{"alias_kind":"pith_short_8","alias_value":"VWUMGNQ5","created_at":"2026-07-05T11:42:19Z"}],"graph_snapshots":[{"event_id":"sha256:e9c4632167dc975a315dc6e19e409d3d5d33ade389bd3574fd4ec192993e897f","target":"graph","created_at":"2026-07-05T11:42:19Z","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/2309.03791/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"We introduce a new class of optimal-transport-regularized divergences, $D^c$, constructed via an infimal convolution between an information divergence, $D$, and an optimal-transport (OT) cost, $C$, and study their use in distributionally robust optimization (DRO). In particular, we propose the $ARMOR_D$ methods as novel approaches to enhancing the adversarial robustness of deep learning models. These DRO-based methods are defined by minimizing the maximum expected loss over a $D^c$-neighborhood of the empirical distribution of the training data. Viewed as a tool for constructing adversarial sa","authors_text":"Jeremiah Birrell, Reza Ebrahimi","cross_cats":["stat.ML"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2023-09-07T15:41:45Z","title":"Optimal Transport Regularized Divergences: Application to Adversarial Robustness"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2309.03791","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":"sha256:245993df9ef5a7778afd45c9bd5b51b1eb934b731ea59611d357423f7fdf1cd2","target":"record","created_at":"2026-07-05T11:42:19Z","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":"1b867e5c899950286e67e34ecef2388f831e01c52765ae15f75072fb7c952871","cross_cats_sorted":["stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2023-09-07T15:41:45Z","title_canon_sha256":"34b9da94183514ae4171c9d79b88ee9e83b2bb15634bf1c68b5b4960a09a84f2"},"schema_version":"1.0","source":{"id":"2309.03791","kind":"arxiv","version":3}},"canonical_sha256":"ada8c3361d328a36ed05b4fa12b3a2656832ca0d4d2db17f672d004f1209f366","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"ada8c3361d328a36ed05b4fa12b3a2656832ca0d4d2db17f672d004f1209f366","first_computed_at":"2026-07-05T11:42:19.812231Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:42:19.812231Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"VHbUiL3kLq7Z4meyeyjkl7Das2Qibl1SbyN+0hk9DFmYfNveN+M4STwwtpQYn8SeCOe6f0t8TY09ZNe5alU2Aw==","signature_status":"signed_v1","signed_at":"2026-07-05T11:42:19.812644Z","signed_message":"canonical_sha256_bytes"},"source_id":"2309.03791","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:245993df9ef5a7778afd45c9bd5b51b1eb934b731ea59611d357423f7fdf1cd2","sha256:e9c4632167dc975a315dc6e19e409d3d5d33ade389bd3574fd4ec192993e897f"],"state_sha256":"fa8918d84b95e291a0dde42daaae388f69a321eb3c1743e8a887ce803119b1b0"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"7lNs/JyPJB7f58jcW4PQ5JeRXWVgQL7BF8LTUQdUBD4ONc2qKlUd/cjoAMYoxgS7wnK9RaCQvdlgf7YU+viDAg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-09T04:36:46.875156Z","bundle_sha256":"5f651ed733af6a19d4b5b83784aa8428053d85e0d0558fd805759af57f9846d4"}}