{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:HGRWZJFKYGJJESIMTTKW4R3KQB","short_pith_number":"pith:HGRWZJFK","canonical_record":{"source":{"id":"2509.06264","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2025-09-08T01:06:45Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"1e2fa10cede5e8d98865b0a7ad3e3c4d94aa3909708142ef85de579afdd41825","abstract_canon_sha256":"259021a8e62f8386c5c40bc84a77039b3f1d32ee09eaabf13c1f6f89171082fa"},"schema_version":"1.0"},"canonical_sha256":"39a36ca4aac19292490c9cd56e476a8046f1c85a095180a25d2cf92a12489807","source":{"kind":"arxiv","id":"2509.06264","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2509.06264","created_at":"2026-07-05T12:06:32Z"},{"alias_kind":"arxiv_version","alias_value":"2509.06264v1","created_at":"2026-07-05T12:06:32Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2509.06264","created_at":"2026-07-05T12:06:32Z"},{"alias_kind":"pith_short_12","alias_value":"HGRWZJFKYGJJ","created_at":"2026-07-05T12:06:32Z"},{"alias_kind":"pith_short_16","alias_value":"HGRWZJFKYGJJESIM","created_at":"2026-07-05T12:06:32Z"},{"alias_kind":"pith_short_8","alias_value":"HGRWZJFK","created_at":"2026-07-05T12:06:32Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:HGRWZJFKYGJJESIMTTKW4R3KQB","target":"record","payload":{"canonical_record":{"source":{"id":"2509.06264","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2025-09-08T01:06:45Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"1e2fa10cede5e8d98865b0a7ad3e3c4d94aa3909708142ef85de579afdd41825","abstract_canon_sha256":"259021a8e62f8386c5c40bc84a77039b3f1d32ee09eaabf13c1f6f89171082fa"},"schema_version":"1.0"},"canonical_sha256":"39a36ca4aac19292490c9cd56e476a8046f1c85a095180a25d2cf92a12489807","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T12:06:32.607522Z","signature_b64":"M2iyQcT4yVVbEIkm2Wb6cQ0FzHmKO34wpvzLAY95imye2qL3DDUlPE1UVK8VrMTIzZ8AAQ+p2oZNwZ3jU3gYAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"39a36ca4aac19292490c9cd56e476a8046f1c85a095180a25d2cf92a12489807","last_reissued_at":"2026-07-05T12:06:32.606946Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T12:06:32.606946Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2509.06264","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":"2026-07-05T12:06:32Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"7N9X6VvYCDt+sfz3E6UQmCJPDurQ4DI3q+4S3Pu2/A93Qw6OKXJHAxtT7laF6cM5oYIJA/ToStAJRIOe9sd5Bg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-05T16:37:07.172874Z"},"content_sha256":"9c63cee39a25b718598cd6c403fb6dd525676b3daa5a507683d65885442e482f","schema_version":"1.0","event_id":"sha256:9c63cee39a25b718598cd6c403fb6dd525676b3daa5a507683d65885442e482f"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:HGRWZJFKYGJJESIMTTKW4R3KQB","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"PLRV-O: Advancing Differentially Private Deep Learning via Privacy Loss Random Variable Optimization","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CR","authors_text":"Ashish Kundu, Ayesha Samreen, Christopher J Quinn, Han Wang, Meisam Mohammady, Nicholas Stout, Qin Yang, Yan Yan, Yuan Hong","submitted_at":"2025-09-08T01:06:45Z","abstract_excerpt":"Differentially Private Stochastic Gradient Descent (DP-SGD) is a standard method for enforcing privacy in deep learning, typically using the Gaussian mechanism to perturb gradient updates. However, conventional mechanisms such as Gaussian and Laplacian noise are parameterized only by variance or scale. This single degree of freedom ties the magnitude of noise directly to both privacy loss and utility degradation, preventing independent control of these two factors. The problem becomes more pronounced when the number of composition rounds T and batch size B vary across tasks, as these variation"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2509.06264","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/2509.06264/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":"2026-07-05T12:06:32Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"nOcVRyINM53NJ7KFCkvkh8sUuhwU1evz1AOx/+I1K1eTTq+yFWzv7MnpJP0PL3NygmoEvKmkgCQ37eOLPSJJDQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-05T16:37:07.173761Z"},"content_sha256":"217175fe388e3fede2f3686850e54a4ba25aac16120e904e271380aa14d4bf20","schema_version":"1.0","event_id":"sha256:217175fe388e3fede2f3686850e54a4ba25aac16120e904e271380aa14d4bf20"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/HGRWZJFKYGJJESIMTTKW4R3KQB/bundle.json","state_url":"https://pith.science/pith/HGRWZJFKYGJJESIMTTKW4R3KQB/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/HGRWZJFKYGJJESIMTTKW4R3KQB/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":"2026-08-05T16:37:07Z","links":{"resolver":"https://pith.science/pith/HGRWZJFKYGJJESIMTTKW4R3KQB","bundle":"https://pith.science/pith/HGRWZJFKYGJJESIMTTKW4R3KQB/bundle.json","state":"https://pith.science/pith/HGRWZJFKYGJJESIMTTKW4R3KQB/state.json","well_known_bundle":"https://pith.science/.well-known/pith/HGRWZJFKYGJJESIMTTKW4R3KQB/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:HGRWZJFKYGJJESIMTTKW4R3KQB","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":"259021a8e62f8386c5c40bc84a77039b3f1d32ee09eaabf13c1f6f89171082fa","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2025-09-08T01:06:45Z","title_canon_sha256":"1e2fa10cede5e8d98865b0a7ad3e3c4d94aa3909708142ef85de579afdd41825"},"schema_version":"1.0","source":{"id":"2509.06264","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2509.06264","created_at":"2026-07-05T12:06:32Z"},{"alias_kind":"arxiv_version","alias_value":"2509.06264v1","created_at":"2026-07-05T12:06:32Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2509.06264","created_at":"2026-07-05T12:06:32Z"},{"alias_kind":"pith_short_12","alias_value":"HGRWZJFKYGJJ","created_at":"2026-07-05T12:06:32Z"},{"alias_kind":"pith_short_16","alias_value":"HGRWZJFKYGJJESIM","created_at":"2026-07-05T12:06:32Z"},{"alias_kind":"pith_short_8","alias_value":"HGRWZJFK","created_at":"2026-07-05T12:06:32Z"}],"graph_snapshots":[{"event_id":"sha256:217175fe388e3fede2f3686850e54a4ba25aac16120e904e271380aa14d4bf20","target":"graph","created_at":"2026-07-05T12:06:32Z","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/2509.06264/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Differentially Private Stochastic Gradient Descent (DP-SGD) is a standard method for enforcing privacy in deep learning, typically using the Gaussian mechanism to perturb gradient updates. However, conventional mechanisms such as Gaussian and Laplacian noise are parameterized only by variance or scale. This single degree of freedom ties the magnitude of noise directly to both privacy loss and utility degradation, preventing independent control of these two factors. The problem becomes more pronounced when the number of composition rounds T and batch size B vary across tasks, as these variation","authors_text":"Ashish Kundu, Ayesha Samreen, Christopher J Quinn, Han Wang, Meisam Mohammady, Nicholas Stout, Qin Yang, Yan Yan, Yuan Hong","cross_cats":["cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2025-09-08T01:06:45Z","title":"PLRV-O: Advancing Differentially Private Deep Learning via Privacy Loss Random Variable Optimization"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2509.06264","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":"sha256:9c63cee39a25b718598cd6c403fb6dd525676b3daa5a507683d65885442e482f","target":"record","created_at":"2026-07-05T12:06:32Z","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":"259021a8e62f8386c5c40bc84a77039b3f1d32ee09eaabf13c1f6f89171082fa","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2025-09-08T01:06:45Z","title_canon_sha256":"1e2fa10cede5e8d98865b0a7ad3e3c4d94aa3909708142ef85de579afdd41825"},"schema_version":"1.0","source":{"id":"2509.06264","kind":"arxiv","version":1}},"canonical_sha256":"39a36ca4aac19292490c9cd56e476a8046f1c85a095180a25d2cf92a12489807","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"39a36ca4aac19292490c9cd56e476a8046f1c85a095180a25d2cf92a12489807","first_computed_at":"2026-07-05T12:06:32.606946Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T12:06:32.606946Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"M2iyQcT4yVVbEIkm2Wb6cQ0FzHmKO34wpvzLAY95imye2qL3DDUlPE1UVK8VrMTIzZ8AAQ+p2oZNwZ3jU3gYAQ==","signature_status":"signed_v1","signed_at":"2026-07-05T12:06:32.607522Z","signed_message":"canonical_sha256_bytes"},"source_id":"2509.06264","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:9c63cee39a25b718598cd6c403fb6dd525676b3daa5a507683d65885442e482f","sha256:217175fe388e3fede2f3686850e54a4ba25aac16120e904e271380aa14d4bf20"],"state_sha256":"e834a662498962298d607d673c01aaaa90c9023bbbd28848a4d1ce3581e6ae34"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"l2CIfAphx5E3hFLsWcDO4xvHxnXgnYc74aotXRNxqDKFPjcGJJyYlOOJ2GiYdETMRQ2GVjNGERTbpxW6kbuaCQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-05T16:37:07.184958Z","bundle_sha256":"00b9c42cf96bf014a8648ad36c74d27b6f34e6e9c130637075b453552d8a202e"}}