{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2018:ON2NC7OPNXMFSQQWW2WKXVV2SN","short_pith_number":"pith:ON2NC7OP","canonical_record":{"source":{"id":"1807.09659","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-07-25T15:20:02Z","cross_cats_sorted":["cs.AI","cs.CV","stat.ML"],"title_canon_sha256":"c65ad719f9845f7adb5423cd94f239442362123b2ebbee5990caa5f761e67e43","abstract_canon_sha256":"cdb1cfc95873ebe171059085ee3ef295a84ee0598941a8611d4ba60819f24f3e"},"schema_version":"1.0"},"canonical_sha256":"7374d17dcf6dd8594216b6acabd6ba93584294318b2eea5c09d328bde1862443","source":{"kind":"arxiv","id":"1807.09659","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1807.09659","created_at":"2026-05-18T00:09:50Z"},{"alias_kind":"arxiv_version","alias_value":"1807.09659v1","created_at":"2026-05-18T00:09:50Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1807.09659","created_at":"2026-05-18T00:09:50Z"},{"alias_kind":"pith_short_12","alias_value":"ON2NC7OPNXMF","created_at":"2026-05-18T12:32:43Z"},{"alias_kind":"pith_short_16","alias_value":"ON2NC7OPNXMFSQQW","created_at":"2026-05-18T12:32:43Z"},{"alias_kind":"pith_short_8","alias_value":"ON2NC7OP","created_at":"2026-05-18T12:32:43Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2018:ON2NC7OPNXMFSQQWW2WKXVV2SN","target":"record","payload":{"canonical_record":{"source":{"id":"1807.09659","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-07-25T15:20:02Z","cross_cats_sorted":["cs.AI","cs.CV","stat.ML"],"title_canon_sha256":"c65ad719f9845f7adb5423cd94f239442362123b2ebbee5990caa5f761e67e43","abstract_canon_sha256":"cdb1cfc95873ebe171059085ee3ef295a84ee0598941a8611d4ba60819f24f3e"},"schema_version":"1.0"},"canonical_sha256":"7374d17dcf6dd8594216b6acabd6ba93584294318b2eea5c09d328bde1862443","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T00:09:50.660222Z","signature_b64":"ip+WL1TxmVg3x695zTkcHKqN7hEVKfvIFj+BPGaqvtWvYC27GO5/8hgtJYIh0wLH3C0AeGf3dvUX19FVdSDdDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7374d17dcf6dd8594216b6acabd6ba93584294318b2eea5c09d328bde1862443","last_reissued_at":"2026-05-18T00:09:50.659685Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T00:09:50.659685Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"1807.09659","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":"2026-05-18T00:09:50Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"oQSARgysyAQMQBh3Lb79CJ9Km/SMx6/CKx4ursxmRT4TKYSPbA/2eqAJdwd2T699j6HlwdzRI+pkKKrGetgnBA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-30T20:54:43.771214Z"},"content_sha256":"485d354749bf929df54dee9947a61ba3e0e3908cc20bfeff8469a68a0ebcf1a3","schema_version":"1.0","event_id":"sha256:485d354749bf929df54dee9947a61ba3e0e3908cc20bfeff8469a68a0ebcf1a3"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2018:ON2NC7OPNXMFSQQWW2WKXVV2SN","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"A Surprising Linear Relationship Predicts Test Performance in Deep Networks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CV","stat.ML"],"primary_cat":"cs.LG","authors_text":"Andrzej Banburski, Brando Miranda, Jack Hidary, Qianli Liao, Tomaso Poggio","submitted_at":"2018-07-25T15:20:02Z","abstract_excerpt":"Given two networks with the same training loss on a dataset, when would they have drastically different test losses and errors? Better understanding of this question of generalization may improve practical applications of deep networks. In this paper we show that with cross-entropy loss it is surprisingly simple to induce significantly different generalization performances for two networks that have the same architecture, the same meta parameters and the same training error: one can either pretrain the networks with different levels of \"corrupted\" data or simply initialize the networks with we"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1807.09659","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":""},"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":"2026-05-18T00:09:50Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"7/d4B+eeY08XLOdGA4FTa9HQ1Zm2yQWSQXvlenl+LXZrv99eJcaEsNlWoZ51GymDieqhEHUCRRPrxscGWx0gBA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-30T20:54:43.771591Z"},"content_sha256":"e65be827a237477b59e47cdfdcd8d20dcd56b71d787bfba484fdc096b94864a2","schema_version":"1.0","event_id":"sha256:e65be827a237477b59e47cdfdcd8d20dcd56b71d787bfba484fdc096b94864a2"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/ON2NC7OPNXMFSQQWW2WKXVV2SN/bundle.json","state_url":"https://pith.science/pith/ON2NC7OPNXMFSQQWW2WKXVV2SN/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/ON2NC7OPNXMFSQQWW2WKXVV2SN/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":"2026-05-30T20:54:43Z","links":{"resolver":"https://pith.science/pith/ON2NC7OPNXMFSQQWW2WKXVV2SN","bundle":"https://pith.science/pith/ON2NC7OPNXMFSQQWW2WKXVV2SN/bundle.json","state":"https://pith.science/pith/ON2NC7OPNXMFSQQWW2WKXVV2SN/state.json","well_known_bundle":"https://pith.science/.well-known/pith/ON2NC7OPNXMFSQQWW2WKXVV2SN/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2018:ON2NC7OPNXMFSQQWW2WKXVV2SN","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":"cdb1cfc95873ebe171059085ee3ef295a84ee0598941a8611d4ba60819f24f3e","cross_cats_sorted":["cs.AI","cs.CV","stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-07-25T15:20:02Z","title_canon_sha256":"c65ad719f9845f7adb5423cd94f239442362123b2ebbee5990caa5f761e67e43"},"schema_version":"1.0","source":{"id":"1807.09659","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1807.09659","created_at":"2026-05-18T00:09:50Z"},{"alias_kind":"arxiv_version","alias_value":"1807.09659v1","created_at":"2026-05-18T00:09:50Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1807.09659","created_at":"2026-05-18T00:09:50Z"},{"alias_kind":"pith_short_12","alias_value":"ON2NC7OPNXMF","created_at":"2026-05-18T12:32:43Z"},{"alias_kind":"pith_short_16","alias_value":"ON2NC7OPNXMFSQQW","created_at":"2026-05-18T12:32:43Z"},{"alias_kind":"pith_short_8","alias_value":"ON2NC7OP","created_at":"2026-05-18T12:32:43Z"}],"graph_snapshots":[{"event_id":"sha256:e65be827a237477b59e47cdfdcd8d20dcd56b71d787bfba484fdc096b94864a2","target":"graph","created_at":"2026-05-18T00:09:50Z","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":"Given two networks with the same training loss on a dataset, when would they have drastically different test losses and errors? Better understanding of this question of generalization may improve practical applications of deep networks. In this paper we show that with cross-entropy loss it is surprisingly simple to induce significantly different generalization performances for two networks that have the same architecture, the same meta parameters and the same training error: one can either pretrain the networks with different levels of \"corrupted\" data or simply initialize the networks with we","authors_text":"Andrzej Banburski, Brando Miranda, Jack Hidary, Qianli Liao, Tomaso Poggio","cross_cats":["cs.AI","cs.CV","stat.ML"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-07-25T15:20:02Z","title":"A Surprising Linear Relationship Predicts Test Performance in Deep Networks"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1807.09659","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":"sha256:485d354749bf929df54dee9947a61ba3e0e3908cc20bfeff8469a68a0ebcf1a3","target":"record","created_at":"2026-05-18T00:09:50Z","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":"cdb1cfc95873ebe171059085ee3ef295a84ee0598941a8611d4ba60819f24f3e","cross_cats_sorted":["cs.AI","cs.CV","stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-07-25T15:20:02Z","title_canon_sha256":"c65ad719f9845f7adb5423cd94f239442362123b2ebbee5990caa5f761e67e43"},"schema_version":"1.0","source":{"id":"1807.09659","kind":"arxiv","version":1}},"canonical_sha256":"7374d17dcf6dd8594216b6acabd6ba93584294318b2eea5c09d328bde1862443","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"7374d17dcf6dd8594216b6acabd6ba93584294318b2eea5c09d328bde1862443","first_computed_at":"2026-05-18T00:09:50.659685Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-18T00:09:50.659685Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"ip+WL1TxmVg3x695zTkcHKqN7hEVKfvIFj+BPGaqvtWvYC27GO5/8hgtJYIh0wLH3C0AeGf3dvUX19FVdSDdDg==","signature_status":"signed_v1","signed_at":"2026-05-18T00:09:50.660222Z","signed_message":"canonical_sha256_bytes"},"source_id":"1807.09659","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:485d354749bf929df54dee9947a61ba3e0e3908cc20bfeff8469a68a0ebcf1a3","sha256:e65be827a237477b59e47cdfdcd8d20dcd56b71d787bfba484fdc096b94864a2"],"state_sha256":"e2c67d952af9548ebbb51e4e014a10994c75af772f2e524714ee0fc2273caebb"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"OwGMwBVHatjjYsDaKYLNp5FLfMerwgndgKdzH7DbSUBaf9I2RdL4c5pBlj8r2CYENI8g4AwhVlkinVrsp11GDQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-30T20:54:43.773517Z","bundle_sha256":"6835db59b0aec0bec17bc7dbc5e957e572de9917f34a33122316ffe3b7274c69"}}