{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2016:3PKSOQVKEVZ6XS2TYDDI5J52QI","short_pith_number":"pith:3PKSOQVK","canonical_record":{"source":{"id":"1602.02389","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2016-02-07T16:50:14Z","cross_cats_sorted":["cs.CV","stat.ML"],"title_canon_sha256":"5ec39f5a59d8c1c4d4e1625cede7790d221e0724cba21de254904dc368cdb612","abstract_canon_sha256":"1fdc49c26320471ef417e0596bee538a9ad193037f3b3ed50bce0599f7a27942"},"schema_version":"1.0"},"canonical_sha256":"dbd52742aa2573ebcb53c0c68ea7ba822f5aa815cd569e15bf14fcea356bd519","source":{"kind":"arxiv","id":"1602.02389","version":4},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1602.02389","created_at":"2026-05-18T00:31:22Z"},{"alias_kind":"arxiv_version","alias_value":"1602.02389v4","created_at":"2026-05-18T00:31:22Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1602.02389","created_at":"2026-05-18T00:31:22Z"},{"alias_kind":"pith_short_12","alias_value":"3PKSOQVKEVZ6","created_at":"2026-05-18T12:29:55Z"},{"alias_kind":"pith_short_16","alias_value":"3PKSOQVKEVZ6XS2T","created_at":"2026-05-18T12:29:55Z"},{"alias_kind":"pith_short_8","alias_value":"3PKSOQVK","created_at":"2026-05-18T12:29:55Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2016:3PKSOQVKEVZ6XS2TYDDI5J52QI","target":"record","payload":{"canonical_record":{"source":{"id":"1602.02389","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2016-02-07T16:50:14Z","cross_cats_sorted":["cs.CV","stat.ML"],"title_canon_sha256":"5ec39f5a59d8c1c4d4e1625cede7790d221e0724cba21de254904dc368cdb612","abstract_canon_sha256":"1fdc49c26320471ef417e0596bee538a9ad193037f3b3ed50bce0599f7a27942"},"schema_version":"1.0"},"canonical_sha256":"dbd52742aa2573ebcb53c0c68ea7ba822f5aa815cd569e15bf14fcea356bd519","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T00:31:22.719455Z","signature_b64":"doWsm8/rWPheYBUWZwrW9wuSH50aN+5KpnXib0KmrhJ7z6HjoXu7HclM1Lk90o+srtL1tGAWTMBtw9YbuPFaAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"dbd52742aa2573ebcb53c0c68ea7ba822f5aa815cd569e15bf14fcea356bd519","last_reissued_at":"2026-05-18T00:31:22.718952Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T00:31:22.718952Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"1602.02389","source_version":4,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-18T00:31:22Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"T+KxMiaRagqOdpcpPo7croCQbYYEJWd5715gR+51QEkB5ns4EdLr2rRjBBu4asYwZeKpnhp7ZZ0XjENSvjPGDg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-31T03:38:57.713471Z"},"content_sha256":"ecead57531d7504b7f309c7ea9dfcd81ce836cce6cc58b16c8fbead5ef6cb1bc","schema_version":"1.0","event_id":"sha256:ecead57531d7504b7f309c7ea9dfcd81ce836cce6cc58b16c8fbead5ef6cb1bc"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2016:3PKSOQVKEVZ6XS2TYDDI5J52QI","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Ensemble Robustness and Generalization of Stochastic Deep Learning Algorithms","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CV","stat.ML"],"primary_cat":"cs.LG","authors_text":"Alex Sivak, Bingyi Kang, Huan Xu, Jiashi Feng, Shie Mannor, Tom Zahavy","submitted_at":"2016-02-07T16:50:14Z","abstract_excerpt":"The question why deep learning algorithms generalize so well has attracted increasing research interest. However, most of the well-established approaches, such as hypothesis capacity, stability or sparseness, have not provided complete explanations (Zhang et al., 2016; Kawaguchi et al., 2017). In this work, we focus on the robustness approach (Xu & Mannor, 2012), i.e., if the error of a hypothesis will not change much due to perturbations of its training examples, then it will also generalize well. As most deep learning algorithms are stochastic (e.g., Stochastic Gradient Descent, Dropout, and"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1602.02389","kind":"arxiv","version":4},"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":"2026-05-18T00:31:22Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"3CkVNtSLUyVSd0LsuuU1UF3K3h3SVbJmPzjK+/drmKnPIyx+BrBVNbJLbMfcZme12IgYtykwcO5dLy9S1X/JBQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-31T03:38:57.714178Z"},"content_sha256":"b1939eb3f0706cb5134ca96e3c3c86590e3bbd901c2cba013e037ba7464e4c96","schema_version":"1.0","event_id":"sha256:b1939eb3f0706cb5134ca96e3c3c86590e3bbd901c2cba013e037ba7464e4c96"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/3PKSOQVKEVZ6XS2TYDDI5J52QI/bundle.json","state_url":"https://pith.science/pith/3PKSOQVKEVZ6XS2TYDDI5J52QI/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/3PKSOQVKEVZ6XS2TYDDI5J52QI/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":"2026-05-31T03:38:57Z","links":{"resolver":"https://pith.science/pith/3PKSOQVKEVZ6XS2TYDDI5J52QI","bundle":"https://pith.science/pith/3PKSOQVKEVZ6XS2TYDDI5J52QI/bundle.json","state":"https://pith.science/pith/3PKSOQVKEVZ6XS2TYDDI5J52QI/state.json","well_known_bundle":"https://pith.science/.well-known/pith/3PKSOQVKEVZ6XS2TYDDI5J52QI/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2016:3PKSOQVKEVZ6XS2TYDDI5J52QI","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":"1fdc49c26320471ef417e0596bee538a9ad193037f3b3ed50bce0599f7a27942","cross_cats_sorted":["cs.CV","stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2016-02-07T16:50:14Z","title_canon_sha256":"5ec39f5a59d8c1c4d4e1625cede7790d221e0724cba21de254904dc368cdb612"},"schema_version":"1.0","source":{"id":"1602.02389","kind":"arxiv","version":4}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1602.02389","created_at":"2026-05-18T00:31:22Z"},{"alias_kind":"arxiv_version","alias_value":"1602.02389v4","created_at":"2026-05-18T00:31:22Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1602.02389","created_at":"2026-05-18T00:31:22Z"},{"alias_kind":"pith_short_12","alias_value":"3PKSOQVKEVZ6","created_at":"2026-05-18T12:29:55Z"},{"alias_kind":"pith_short_16","alias_value":"3PKSOQVKEVZ6XS2T","created_at":"2026-05-18T12:29:55Z"},{"alias_kind":"pith_short_8","alias_value":"3PKSOQVK","created_at":"2026-05-18T12:29:55Z"}],"graph_snapshots":[{"event_id":"sha256:b1939eb3f0706cb5134ca96e3c3c86590e3bbd901c2cba013e037ba7464e4c96","target":"graph","created_at":"2026-05-18T00:31:22Z","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":"The question why deep learning algorithms generalize so well has attracted increasing research interest. However, most of the well-established approaches, such as hypothesis capacity, stability or sparseness, have not provided complete explanations (Zhang et al., 2016; Kawaguchi et al., 2017). In this work, we focus on the robustness approach (Xu & Mannor, 2012), i.e., if the error of a hypothesis will not change much due to perturbations of its training examples, then it will also generalize well. As most deep learning algorithms are stochastic (e.g., Stochastic Gradient Descent, Dropout, and","authors_text":"Alex Sivak, Bingyi Kang, Huan Xu, Jiashi Feng, Shie Mannor, Tom Zahavy","cross_cats":["cs.CV","stat.ML"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2016-02-07T16:50:14Z","title":"Ensemble Robustness and Generalization of Stochastic Deep Learning Algorithms"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1602.02389","kind":"arxiv","version":4},"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:ecead57531d7504b7f309c7ea9dfcd81ce836cce6cc58b16c8fbead5ef6cb1bc","target":"record","created_at":"2026-05-18T00:31:22Z","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":"1fdc49c26320471ef417e0596bee538a9ad193037f3b3ed50bce0599f7a27942","cross_cats_sorted":["cs.CV","stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2016-02-07T16:50:14Z","title_canon_sha256":"5ec39f5a59d8c1c4d4e1625cede7790d221e0724cba21de254904dc368cdb612"},"schema_version":"1.0","source":{"id":"1602.02389","kind":"arxiv","version":4}},"canonical_sha256":"dbd52742aa2573ebcb53c0c68ea7ba822f5aa815cd569e15bf14fcea356bd519","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"dbd52742aa2573ebcb53c0c68ea7ba822f5aa815cd569e15bf14fcea356bd519","first_computed_at":"2026-05-18T00:31:22.718952Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-18T00:31:22.718952Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"doWsm8/rWPheYBUWZwrW9wuSH50aN+5KpnXib0KmrhJ7z6HjoXu7HclM1Lk90o+srtL1tGAWTMBtw9YbuPFaAQ==","signature_status":"signed_v1","signed_at":"2026-05-18T00:31:22.719455Z","signed_message":"canonical_sha256_bytes"},"source_id":"1602.02389","source_kind":"arxiv","source_version":4}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:ecead57531d7504b7f309c7ea9dfcd81ce836cce6cc58b16c8fbead5ef6cb1bc","sha256:b1939eb3f0706cb5134ca96e3c3c86590e3bbd901c2cba013e037ba7464e4c96"],"state_sha256":"de16382dae84a44b3c0755448434e1b3ead9c13fd4c71bfba7dd2a29231497f7"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"qABLnxJiX5LJvaL1JB5MiAd0Rw17f2iBOtUp3PDMWXlHHmmd5xEOfB8x4o5iYRFQ0GjdqDirT47Rx0gHkmwVDA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-31T03:38:57.717557Z","bundle_sha256":"1a559be4415f4536e9257baf8246b32b32491b7e3b44cab99ea078c7e0dab1fe"}}