{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2018:4JV6JJL6GLJAQ5EEH57GTP4VK4","short_pith_number":"pith:4JV6JJL6","canonical_record":{"source":{"id":"1803.00404","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2018-02-23T05:02:59Z","cross_cats_sorted":["cs.LG","cs.NE"],"title_canon_sha256":"81bfbc54e93ce71421c096f0e6613d4276244dfd9b3b39ca1c4759d76fbfdf51","abstract_canon_sha256":"5af3c6a0544b835e2d54d2e07d487472a12e19bb6ddbbca3109494f2e5fe43e5"},"schema_version":"1.0"},"canonical_sha256":"e26be4a57e32d20874843f7e69bf95573e81fb64327392627d1ceeb08c8d17b3","source":{"kind":"arxiv","id":"1803.00404","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1803.00404","created_at":"2026-05-17T23:57:54Z"},{"alias_kind":"arxiv_version","alias_value":"1803.00404v3","created_at":"2026-05-17T23:57:54Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1803.00404","created_at":"2026-05-17T23:57:54Z"},{"alias_kind":"pith_short_12","alias_value":"4JV6JJL6GLJA","created_at":"2026-05-18T12:32:05Z"},{"alias_kind":"pith_short_16","alias_value":"4JV6JJL6GLJAQ5EE","created_at":"2026-05-18T12:32:05Z"},{"alias_kind":"pith_short_8","alias_value":"4JV6JJL6","created_at":"2026-05-18T12:32:05Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2018:4JV6JJL6GLJAQ5EEH57GTP4VK4","target":"record","payload":{"canonical_record":{"source":{"id":"1803.00404","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2018-02-23T05:02:59Z","cross_cats_sorted":["cs.LG","cs.NE"],"title_canon_sha256":"81bfbc54e93ce71421c096f0e6613d4276244dfd9b3b39ca1c4759d76fbfdf51","abstract_canon_sha256":"5af3c6a0544b835e2d54d2e07d487472a12e19bb6ddbbca3109494f2e5fe43e5"},"schema_version":"1.0"},"canonical_sha256":"e26be4a57e32d20874843f7e69bf95573e81fb64327392627d1ceeb08c8d17b3","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:57:54.196835Z","signature_b64":"YBEtyh6N6/CbnIsYha7WK98bR7i0J7yOr0qd0xBETQtxwmw043Vs4DLq+A1E61KFyTScl2V0D6rzKtsTk4ZBBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e26be4a57e32d20874843f7e69bf95573e81fb64327392627d1ceeb08c8d17b3","last_reissued_at":"2026-05-17T23:57:54.196276Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:57:54.196276Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"1803.00404","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":"2026-05-17T23:57:54Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"sw5Wl5BcQT6dppfUe59B8FVixiJ9Tulvs4Fh/TekkpK93aE6QxEVRQShtLPw/knxGVbrf22CIMgNtu0hKAcMDw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-27T23:47:57.720609Z"},"content_sha256":"a1bbb6a43e7fed99f7f21451e34e78b8da86dfda274a05bf7ac63a4d1e5c5eee","schema_version":"1.0","event_id":"sha256:a1bbb6a43e7fed99f7f21451e34e78b8da86dfda274a05bf7ac63a4d1e5c5eee"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2018:4JV6JJL6GLJAQ5EEH57GTP4VK4","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Deep Defense: Training DNNs with Improved Adversarial Robustness","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","cs.NE"],"primary_cat":"cs.CV","authors_text":"Changshui Zhang, Yiwen Guo, Ziang Yan","submitted_at":"2018-02-23T05:02:59Z","abstract_excerpt":"Despite the efficacy on a variety of computer vision tasks, deep neural networks (DNNs) are vulnerable to adversarial attacks, limiting their applications in security-critical systems. Recent works have shown the possibility of generating imperceptibly perturbed image inputs (a.k.a., adversarial examples) to fool well-trained DNN classifiers into making arbitrary predictions. To address this problem, we propose a training recipe named \"deep defense\". Our core idea is to integrate an adversarial perturbation-based regularizer into the classification objective, such that the obtained models lear"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1803.00404","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":""},"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":"2026-05-17T23:57:54Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"e5bqVAMy92ROW2FkuWm7VBmxERLk9Vc07UsGb8o2qUrvIWwfuN0sjBkHebVxobFLgY+iGLa4d/krEx0oP9uhBw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-27T23:47:57.721402Z"},"content_sha256":"4ce349e4990df661546e66294a5ae7527e2e62d35aad01839e5bc4ad8a52a130","schema_version":"1.0","event_id":"sha256:4ce349e4990df661546e66294a5ae7527e2e62d35aad01839e5bc4ad8a52a130"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/4JV6JJL6GLJAQ5EEH57GTP4VK4/bundle.json","state_url":"https://pith.science/pith/4JV6JJL6GLJAQ5EEH57GTP4VK4/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/4JV6JJL6GLJAQ5EEH57GTP4VK4/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":"2026-05-27T23:47:57Z","links":{"resolver":"https://pith.science/pith/4JV6JJL6GLJAQ5EEH57GTP4VK4","bundle":"https://pith.science/pith/4JV6JJL6GLJAQ5EEH57GTP4VK4/bundle.json","state":"https://pith.science/pith/4JV6JJL6GLJAQ5EEH57GTP4VK4/state.json","well_known_bundle":"https://pith.science/.well-known/pith/4JV6JJL6GLJAQ5EEH57GTP4VK4/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2018:4JV6JJL6GLJAQ5EEH57GTP4VK4","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":"5af3c6a0544b835e2d54d2e07d487472a12e19bb6ddbbca3109494f2e5fe43e5","cross_cats_sorted":["cs.LG","cs.NE"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2018-02-23T05:02:59Z","title_canon_sha256":"81bfbc54e93ce71421c096f0e6613d4276244dfd9b3b39ca1c4759d76fbfdf51"},"schema_version":"1.0","source":{"id":"1803.00404","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1803.00404","created_at":"2026-05-17T23:57:54Z"},{"alias_kind":"arxiv_version","alias_value":"1803.00404v3","created_at":"2026-05-17T23:57:54Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1803.00404","created_at":"2026-05-17T23:57:54Z"},{"alias_kind":"pith_short_12","alias_value":"4JV6JJL6GLJA","created_at":"2026-05-18T12:32:05Z"},{"alias_kind":"pith_short_16","alias_value":"4JV6JJL6GLJAQ5EE","created_at":"2026-05-18T12:32:05Z"},{"alias_kind":"pith_short_8","alias_value":"4JV6JJL6","created_at":"2026-05-18T12:32:05Z"}],"graph_snapshots":[{"event_id":"sha256:4ce349e4990df661546e66294a5ae7527e2e62d35aad01839e5bc4ad8a52a130","target":"graph","created_at":"2026-05-17T23:57:54Z","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"},"paper":{"abstract_excerpt":"Despite the efficacy on a variety of computer vision tasks, deep neural networks (DNNs) are vulnerable to adversarial attacks, limiting their applications in security-critical systems. Recent works have shown the possibility of generating imperceptibly perturbed image inputs (a.k.a., adversarial examples) to fool well-trained DNN classifiers into making arbitrary predictions. To address this problem, we propose a training recipe named \"deep defense\". Our core idea is to integrate an adversarial perturbation-based regularizer into the classification objective, such that the obtained models lear","authors_text":"Changshui Zhang, Yiwen Guo, Ziang Yan","cross_cats":["cs.LG","cs.NE"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2018-02-23T05:02:59Z","title":"Deep Defense: Training DNNs with Improved Adversarial Robustness"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1803.00404","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:a1bbb6a43e7fed99f7f21451e34e78b8da86dfda274a05bf7ac63a4d1e5c5eee","target":"record","created_at":"2026-05-17T23:57:54Z","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":"5af3c6a0544b835e2d54d2e07d487472a12e19bb6ddbbca3109494f2e5fe43e5","cross_cats_sorted":["cs.LG","cs.NE"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2018-02-23T05:02:59Z","title_canon_sha256":"81bfbc54e93ce71421c096f0e6613d4276244dfd9b3b39ca1c4759d76fbfdf51"},"schema_version":"1.0","source":{"id":"1803.00404","kind":"arxiv","version":3}},"canonical_sha256":"e26be4a57e32d20874843f7e69bf95573e81fb64327392627d1ceeb08c8d17b3","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"e26be4a57e32d20874843f7e69bf95573e81fb64327392627d1ceeb08c8d17b3","first_computed_at":"2026-05-17T23:57:54.196276Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-17T23:57:54.196276Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"YBEtyh6N6/CbnIsYha7WK98bR7i0J7yOr0qd0xBETQtxwmw043Vs4DLq+A1E61KFyTScl2V0D6rzKtsTk4ZBBg==","signature_status":"signed_v1","signed_at":"2026-05-17T23:57:54.196835Z","signed_message":"canonical_sha256_bytes"},"source_id":"1803.00404","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:a1bbb6a43e7fed99f7f21451e34e78b8da86dfda274a05bf7ac63a4d1e5c5eee","sha256:4ce349e4990df661546e66294a5ae7527e2e62d35aad01839e5bc4ad8a52a130"],"state_sha256":"7c8098c49ec3f1eed49f65d879f1dc10c6528bf552bb282cd40947e8d5aaee5c"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Kxvu5R2E6xk01rclb+oCAxT1hUCUNTiyqmTnTIoWuy+2nSf5Tqb4J/pVeMO2NnlxE5gfmfuBIMQcNElcgYK7Cw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-27T23:47:57.725048Z","bundle_sha256":"640c93410ba4478407ef4aa2667313eb73cec156b0c05150b3b4f7de305a8b2f"}}