{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:L6LHFZOWRBZWFMKCDGGIVPSUSB","short_pith_number":"pith:L6LHFZOW","canonical_record":{"source":{"id":"2505.14021","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.LG","submitted_at":"2025-05-20T07:22:21Z","cross_cats_sorted":["cs.CV","stat.ML"],"title_canon_sha256":"529a102feb8e8537126f428885e49b60b60ba8954e7d146d58bf1789c793712d","abstract_canon_sha256":"b7a4129a6644057f59c2b07d26a1bedca5338c1c55ae158b7501ee925ad5a7db"},"schema_version":"1.0"},"canonical_sha256":"5f9672e5d6887362b142198c8abe54906f1dd27579b7ec9787496b5bfbf65118","source":{"kind":"arxiv","id":"2505.14021","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2505.14021","created_at":"2026-07-05T11:05:54Z"},{"alias_kind":"arxiv_version","alias_value":"2505.14021v1","created_at":"2026-07-05T11:05:54Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.14021","created_at":"2026-07-05T11:05:54Z"},{"alias_kind":"pith_short_12","alias_value":"L6LHFZOWRBZW","created_at":"2026-07-05T11:05:54Z"},{"alias_kind":"pith_short_16","alias_value":"L6LHFZOWRBZWFMKC","created_at":"2026-07-05T11:05:54Z"},{"alias_kind":"pith_short_8","alias_value":"L6LHFZOW","created_at":"2026-07-05T11:05:54Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:L6LHFZOWRBZWFMKCDGGIVPSUSB","target":"record","payload":{"canonical_record":{"source":{"id":"2505.14021","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.LG","submitted_at":"2025-05-20T07:22:21Z","cross_cats_sorted":["cs.CV","stat.ML"],"title_canon_sha256":"529a102feb8e8537126f428885e49b60b60ba8954e7d146d58bf1789c793712d","abstract_canon_sha256":"b7a4129a6644057f59c2b07d26a1bedca5338c1c55ae158b7501ee925ad5a7db"},"schema_version":"1.0"},"canonical_sha256":"5f9672e5d6887362b142198c8abe54906f1dd27579b7ec9787496b5bfbf65118","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:05:54.849496Z","signature_b64":"B+9tWgIoYN9lp+kM+Sbqpet28zTEJSQAspSwndRKW0hHM9z77fXC+3Ns5rljUSDRxAe/xSOWZd3Lb+CB3C9EDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"5f9672e5d6887362b142198c8abe54906f1dd27579b7ec9787496b5bfbf65118","last_reissued_at":"2026-07-05T11:05:54.849157Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:05:54.849157Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2505.14021","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":"2026-07-05T11:05:54Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"j5OzmhB8XZhM+LsnAoVam6gLenvxK46J+W4xdDyU6GE4kBFrKN60UF5eRluEEhgrItVa/GjGfHdpsK1/yc6rAQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T08:44:27.088742Z"},"content_sha256":"90f7cc7660620db80589d31ea196128d31fbdc21e18832353a847792ce403e66","schema_version":"1.0","event_id":"sha256:90f7cc7660620db80589d31ea196128d31fbdc21e18832353a847792ce403e66"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:L6LHFZOWRBZWFMKCDGGIVPSUSB","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Adversarial Training from Mean Field Perspective","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.CV","stat.ML"],"primary_cat":"cs.LG","authors_text":"Hiroshi Kera, Soichiro Kumano, Toshihiko Yamasaki","submitted_at":"2025-05-20T07:22:21Z","abstract_excerpt":"Although adversarial training is known to be effective against adversarial examples, training dynamics are not well understood. In this study, we present the first theoretical analysis of adversarial training in random deep neural networks without any assumptions on data distributions. We introduce a new theoretical framework based on mean field theory, which addresses the limitations of existing mean field-based approaches. Based on this framework, we derive (empirically tight) upper bounds of $\\ell_q$ norm-based adversarial loss with $\\ell_p$ norm-based adversarial examples for various value"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.14021","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/2505.14021/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":"2026-07-05T11:05:54Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"LY+v7URfGgZn4D+DPjihq54oviK+C30DMkLdiar9hw5CmjC/j4ePOSnq0SVhDbtnqAr5K4mHxRIXN7w31MM3Bg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T08:44:27.089711Z"},"content_sha256":"7712551c4ec4368e63d844df370c446b0ef0de21ec19de5e82e0a85b1f635f36","schema_version":"1.0","event_id":"sha256:7712551c4ec4368e63d844df370c446b0ef0de21ec19de5e82e0a85b1f635f36"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/L6LHFZOWRBZWFMKCDGGIVPSUSB/bundle.json","state_url":"https://pith.science/pith/L6LHFZOWRBZWFMKCDGGIVPSUSB/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/L6LHFZOWRBZWFMKCDGGIVPSUSB/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":"2026-08-08T08:44:27Z","links":{"resolver":"https://pith.science/pith/L6LHFZOWRBZWFMKCDGGIVPSUSB","bundle":"https://pith.science/pith/L6LHFZOWRBZWFMKCDGGIVPSUSB/bundle.json","state":"https://pith.science/pith/L6LHFZOWRBZWFMKCDGGIVPSUSB/state.json","well_known_bundle":"https://pith.science/.well-known/pith/L6LHFZOWRBZWFMKCDGGIVPSUSB/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:L6LHFZOWRBZWFMKCDGGIVPSUSB","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":"b7a4129a6644057f59c2b07d26a1bedca5338c1c55ae158b7501ee925ad5a7db","cross_cats_sorted":["cs.CV","stat.ML"],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.LG","submitted_at":"2025-05-20T07:22:21Z","title_canon_sha256":"529a102feb8e8537126f428885e49b60b60ba8954e7d146d58bf1789c793712d"},"schema_version":"1.0","source":{"id":"2505.14021","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2505.14021","created_at":"2026-07-05T11:05:54Z"},{"alias_kind":"arxiv_version","alias_value":"2505.14021v1","created_at":"2026-07-05T11:05:54Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.14021","created_at":"2026-07-05T11:05:54Z"},{"alias_kind":"pith_short_12","alias_value":"L6LHFZOWRBZW","created_at":"2026-07-05T11:05:54Z"},{"alias_kind":"pith_short_16","alias_value":"L6LHFZOWRBZWFMKC","created_at":"2026-07-05T11:05:54Z"},{"alias_kind":"pith_short_8","alias_value":"L6LHFZOW","created_at":"2026-07-05T11:05:54Z"}],"graph_snapshots":[{"event_id":"sha256:7712551c4ec4368e63d844df370c446b0ef0de21ec19de5e82e0a85b1f635f36","target":"graph","created_at":"2026-07-05T11:05:54Z","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/2505.14021/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Although adversarial training is known to be effective against adversarial examples, training dynamics are not well understood. In this study, we present the first theoretical analysis of adversarial training in random deep neural networks without any assumptions on data distributions. We introduce a new theoretical framework based on mean field theory, which addresses the limitations of existing mean field-based approaches. Based on this framework, we derive (empirically tight) upper bounds of $\\ell_q$ norm-based adversarial loss with $\\ell_p$ norm-based adversarial examples for various value","authors_text":"Hiroshi Kera, Soichiro Kumano, Toshihiko Yamasaki","cross_cats":["cs.CV","stat.ML"],"headline":"","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.LG","submitted_at":"2025-05-20T07:22:21Z","title":"Adversarial Training from Mean Field Perspective"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.14021","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":"sha256:90f7cc7660620db80589d31ea196128d31fbdc21e18832353a847792ce403e66","target":"record","created_at":"2026-07-05T11:05:54Z","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":"b7a4129a6644057f59c2b07d26a1bedca5338c1c55ae158b7501ee925ad5a7db","cross_cats_sorted":["cs.CV","stat.ML"],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.LG","submitted_at":"2025-05-20T07:22:21Z","title_canon_sha256":"529a102feb8e8537126f428885e49b60b60ba8954e7d146d58bf1789c793712d"},"schema_version":"1.0","source":{"id":"2505.14021","kind":"arxiv","version":1}},"canonical_sha256":"5f9672e5d6887362b142198c8abe54906f1dd27579b7ec9787496b5bfbf65118","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"5f9672e5d6887362b142198c8abe54906f1dd27579b7ec9787496b5bfbf65118","first_computed_at":"2026-07-05T11:05:54.849157Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:05:54.849157Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"B+9tWgIoYN9lp+kM+Sbqpet28zTEJSQAspSwndRKW0hHM9z77fXC+3Ns5rljUSDRxAe/xSOWZd3Lb+CB3C9EDQ==","signature_status":"signed_v1","signed_at":"2026-07-05T11:05:54.849496Z","signed_message":"canonical_sha256_bytes"},"source_id":"2505.14021","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:90f7cc7660620db80589d31ea196128d31fbdc21e18832353a847792ce403e66","sha256:7712551c4ec4368e63d844df370c446b0ef0de21ec19de5e82e0a85b1f635f36"],"state_sha256":"75094d58424a6754035e2d87c545c0405e2daf128b7d25c4b1663e5945d4d3ed"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"XDaED34GUA5y1fFtEiR7fyrLf8B61tesazl/okEnOEjItWd9PkBOa1otJKBUWQOnSo511mzD2poDfbwUQ9ibCQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-08T08:44:27.097331Z","bundle_sha256":"643243d0382f058a94a74ade3d6c1f2c6e5ec95d06379ff04ed9ccb1b11600bb"}}