{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2020:BC5IAEEBC5WPI32GCMOIA4OOCP","short_pith_number":"pith:BC5IAEEB","canonical_record":{"source":{"id":"2008.06631","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"stat.ML","submitted_at":"2020-08-15T02:32:09Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"c835b7275afcf02713696b5e4ea5a753c980c3d3c63e6897ce54389fbe6b7619","abstract_canon_sha256":"1c66813d369ecee0cab17a859506f839f91871478734bba4f653c3f9b79bb2fc"},"schema_version":"1.0"},"canonical_sha256":"08ba801081176cf46f46131c8071ce13cd6a3893cb27d4ede114c3150b7cf1c5","source":{"kind":"arxiv","id":"2008.06631","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2008.06631","created_at":"2026-07-05T02:29:29Z"},{"alias_kind":"arxiv_version","alias_value":"2008.06631v2","created_at":"2026-07-05T02:29:29Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2008.06631","created_at":"2026-07-05T02:29:29Z"},{"alias_kind":"pith_short_12","alias_value":"BC5IAEEBC5WP","created_at":"2026-07-05T02:29:29Z"},{"alias_kind":"pith_short_16","alias_value":"BC5IAEEBC5WPI32G","created_at":"2026-07-05T02:29:29Z"},{"alias_kind":"pith_short_8","alias_value":"BC5IAEEB","created_at":"2026-07-05T02:29:29Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2020:BC5IAEEBC5WPI32GCMOIA4OOCP","target":"record","payload":{"canonical_record":{"source":{"id":"2008.06631","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"stat.ML","submitted_at":"2020-08-15T02:32:09Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"c835b7275afcf02713696b5e4ea5a753c980c3d3c63e6897ce54389fbe6b7619","abstract_canon_sha256":"1c66813d369ecee0cab17a859506f839f91871478734bba4f653c3f9b79bb2fc"},"schema_version":"1.0"},"canonical_sha256":"08ba801081176cf46f46131c8071ce13cd6a3893cb27d4ede114c3150b7cf1c5","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:29:29.893596Z","signature_b64":"vRoES9HW8HM+FWa3BfemdIC+9W+g56rGp6zFTgua6mzwyncYSFMRpqxyyfQreCH6nSFVPstKvAp7Dw72mSe5Bg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"08ba801081176cf46f46131c8071ce13cd6a3893cb27d4ede114c3150b7cf1c5","last_reissued_at":"2026-07-05T02:29:29.893098Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:29:29.893098Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2008.06631","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T02:29:29Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"nT0xRxf/pzHbNClMDdBIkawa8uiCWLXVJTqzGAEGvcwDXKFkAeGeWkQotd88AwCQgLvpjQv1wd2xK9KxdODaDQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-07T08:39:31.404469Z"},"content_sha256":"74069e5e2ac336c06c7c3ba0e65db794f30e352cbf891087a5ef79a869e79f27","schema_version":"1.0","event_id":"sha256:74069e5e2ac336c06c7c3ba0e65db794f30e352cbf891087a5ef79a869e79f27"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2020:BC5IAEEBC5WPI32GCMOIA4OOCP","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"On the Generalization Properties of Adversarial Training","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"stat.ML","authors_text":"Guang Cheng, Qifan Song, Yue Xing","submitted_at":"2020-08-15T02:32:09Z","abstract_excerpt":"Modern machine learning and deep learning models are shown to be vulnerable when testing data are slightly perturbed. Existing theoretical studies of adversarial training algorithms mostly focus on either adversarial training losses or local convergence properties. In contrast, this paper studies the generalization performance of a generic adversarial training algorithm. Specifically, we consider linear regression models and two-layer neural networks (with lazy training) using squared loss under low-dimensional and high-dimensional regimes. In the former regime, after overcoming the non-smooth"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2008.06631","kind":"arxiv","version":2},"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/2008.06631/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":"2026-07-05T02:29:29Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"5muX+m758aUvK+abmjPpPRXFjEulQf0q85bK1dqUxOf3MC5guzheVwCSbVW39O2XX15jTqpX3T9yDGLt2F/4Dg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-07T08:39:31.410700Z"},"content_sha256":"07ab83f12669de7a7de8fde3fedbf6f412d4cdffaea836351d34ff69bdb40020","schema_version":"1.0","event_id":"sha256:07ab83f12669de7a7de8fde3fedbf6f412d4cdffaea836351d34ff69bdb40020"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/BC5IAEEBC5WPI32GCMOIA4OOCP/bundle.json","state_url":"https://pith.science/pith/BC5IAEEBC5WPI32GCMOIA4OOCP/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/BC5IAEEBC5WPI32GCMOIA4OOCP/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":"2026-08-07T08:39:31Z","links":{"resolver":"https://pith.science/pith/BC5IAEEBC5WPI32GCMOIA4OOCP","bundle":"https://pith.science/pith/BC5IAEEBC5WPI32GCMOIA4OOCP/bundle.json","state":"https://pith.science/pith/BC5IAEEBC5WPI32GCMOIA4OOCP/state.json","well_known_bundle":"https://pith.science/.well-known/pith/BC5IAEEBC5WPI32GCMOIA4OOCP/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2020:BC5IAEEBC5WPI32GCMOIA4OOCP","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":"1c66813d369ecee0cab17a859506f839f91871478734bba4f653c3f9b79bb2fc","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"stat.ML","submitted_at":"2020-08-15T02:32:09Z","title_canon_sha256":"c835b7275afcf02713696b5e4ea5a753c980c3d3c63e6897ce54389fbe6b7619"},"schema_version":"1.0","source":{"id":"2008.06631","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2008.06631","created_at":"2026-07-05T02:29:29Z"},{"alias_kind":"arxiv_version","alias_value":"2008.06631v2","created_at":"2026-07-05T02:29:29Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2008.06631","created_at":"2026-07-05T02:29:29Z"},{"alias_kind":"pith_short_12","alias_value":"BC5IAEEBC5WP","created_at":"2026-07-05T02:29:29Z"},{"alias_kind":"pith_short_16","alias_value":"BC5IAEEBC5WPI32G","created_at":"2026-07-05T02:29:29Z"},{"alias_kind":"pith_short_8","alias_value":"BC5IAEEB","created_at":"2026-07-05T02:29:29Z"}],"graph_snapshots":[{"event_id":"sha256:07ab83f12669de7a7de8fde3fedbf6f412d4cdffaea836351d34ff69bdb40020","target":"graph","created_at":"2026-07-05T02:29:29Z","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/2008.06631/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Modern machine learning and deep learning models are shown to be vulnerable when testing data are slightly perturbed. Existing theoretical studies of adversarial training algorithms mostly focus on either adversarial training losses or local convergence properties. In contrast, this paper studies the generalization performance of a generic adversarial training algorithm. Specifically, we consider linear regression models and two-layer neural networks (with lazy training) using squared loss under low-dimensional and high-dimensional regimes. In the former regime, after overcoming the non-smooth","authors_text":"Guang Cheng, Qifan Song, Yue Xing","cross_cats":["cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"stat.ML","submitted_at":"2020-08-15T02:32:09Z","title":"On the Generalization Properties of Adversarial Training"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2008.06631","kind":"arxiv","version":2},"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":"sha256:74069e5e2ac336c06c7c3ba0e65db794f30e352cbf891087a5ef79a869e79f27","target":"record","created_at":"2026-07-05T02:29:29Z","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":"1c66813d369ecee0cab17a859506f839f91871478734bba4f653c3f9b79bb2fc","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"stat.ML","submitted_at":"2020-08-15T02:32:09Z","title_canon_sha256":"c835b7275afcf02713696b5e4ea5a753c980c3d3c63e6897ce54389fbe6b7619"},"schema_version":"1.0","source":{"id":"2008.06631","kind":"arxiv","version":2}},"canonical_sha256":"08ba801081176cf46f46131c8071ce13cd6a3893cb27d4ede114c3150b7cf1c5","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"08ba801081176cf46f46131c8071ce13cd6a3893cb27d4ede114c3150b7cf1c5","first_computed_at":"2026-07-05T02:29:29.893098Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T02:29:29.893098Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"vRoES9HW8HM+FWa3BfemdIC+9W+g56rGp6zFTgua6mzwyncYSFMRpqxyyfQreCH6nSFVPstKvAp7Dw72mSe5Bg==","signature_status":"signed_v1","signed_at":"2026-07-05T02:29:29.893596Z","signed_message":"canonical_sha256_bytes"},"source_id":"2008.06631","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:74069e5e2ac336c06c7c3ba0e65db794f30e352cbf891087a5ef79a869e79f27","sha256:07ab83f12669de7a7de8fde3fedbf6f412d4cdffaea836351d34ff69bdb40020"],"state_sha256":"aeb33f26bc2ef09235695b2ebd1804ef16c7a6c7772999a6790242f0867631e0"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"DvFBE8DsU00skRRqqScP/mKgygSPuwTdCtSlvxPhOBrkUO23C02t4O4Vc3yjAdwUHs5g5cmHdn7l+bOZrTKoAg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-07T08:39:31.422488Z","bundle_sha256":"387b31c6c71b65c8e0f38f3a67008af8322840ec171f051507f55ee46d230571"}}