{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2018:PTZLBULTCN4NJU64CF3XCM4EHP","short_pith_number":"pith:PTZLBULT","schema_version":"1.0","canonical_sha256":"7cf2b0d1731378d4d3dc11777133843bed5a4269db0142fa6b7b41080d3eaa1a","source":{"kind":"arxiv","id":"1806.08805","version":1},"attestation_state":"computed","paper":{"title":"PCA of high dimensional random walks with comparison to neural network training","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"stat.ML","authors_text":"Jascha Sohl-Dickstein, Joseph M. Antognini","submitted_at":"2018-06-22T18:08:20Z","abstract_excerpt":"One technique to visualize the training of neural networks is to perform PCA on the parameters over the course of training and to project to the subspace spanned by the first few PCA components. In this paper we compare this technique to the PCA of a high dimensional random walk. We compute the eigenvalues and eigenvectors of the covariance of the trajectory and prove that in the long trajectory and high dimensional limit most of the variance is in the first few PCA components, and that the projection of the trajectory onto any subspace spanned by PCA components is a Lissajous curve. We genera"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"1806.08805","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"stat.ML","submitted_at":"2018-06-22T18:08:20Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"2b5ac90c6db63adbae0109ddd88e3c10578291e551e6fd05e416f0e428e509b8","abstract_canon_sha256":"0abc7d8c7b056437a8267b6a87f5c1b7360c990ccd3d71026989fc01c92d1396"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T00:12:31.231852Z","signature_b64":"vi5sutuRwQN/31fDfilR4SwVOtGfds99a0dNO21TiO2STnAbC049bDNfimm5Al9EMqSZlQwc+haXcPscxJR2BA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7cf2b0d1731378d4d3dc11777133843bed5a4269db0142fa6b7b41080d3eaa1a","last_reissued_at":"2026-05-18T00:12:31.231240Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T00:12:31.231240Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"PCA of high dimensional random walks with comparison to neural network training","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"stat.ML","authors_text":"Jascha Sohl-Dickstein, Joseph M. Antognini","submitted_at":"2018-06-22T18:08:20Z","abstract_excerpt":"One technique to visualize the training of neural networks is to perform PCA on the parameters over the course of training and to project to the subspace spanned by the first few PCA components. In this paper we compare this technique to the PCA of a high dimensional random walk. We compute the eigenvalues and eigenvectors of the covariance of the trajectory and prove that in the long trajectory and high dimensional limit most of the variance is in the first few PCA components, and that the projection of the trajectory onto any subspace spanned by PCA components is a Lissajous curve. We genera"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1806.08805","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"},"aliases":[{"alias_kind":"arxiv","alias_value":"1806.08805","created_at":"2026-05-18T00:12:31.231380+00:00"},{"alias_kind":"arxiv_version","alias_value":"1806.08805v1","created_at":"2026-05-18T00:12:31.231380+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1806.08805","created_at":"2026-05-18T00:12:31.231380+00:00"},{"alias_kind":"pith_short_12","alias_value":"PTZLBULTCN4N","created_at":"2026-05-18T12:32:46.962924+00:00"},{"alias_kind":"pith_short_16","alias_value":"PTZLBULTCN4NJU64","created_at":"2026-05-18T12:32:46.962924+00:00"},{"alias_kind":"pith_short_8","alias_value":"PTZLBULT","created_at":"2026-05-18T12:32:46.962924+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/PTZLBULTCN4NJU64CF3XCM4EHP","json":"https://pith.science/pith/PTZLBULTCN4NJU64CF3XCM4EHP.json","graph_json":"https://pith.science/api/pith-number/PTZLBULTCN4NJU64CF3XCM4EHP/graph.json","events_json":"https://pith.science/api/pith-number/PTZLBULTCN4NJU64CF3XCM4EHP/events.json","paper":"https://pith.science/paper/PTZLBULT"},"agent_actions":{"view_html":"https://pith.science/pith/PTZLBULTCN4NJU64CF3XCM4EHP","download_json":"https://pith.science/pith/PTZLBULTCN4NJU64CF3XCM4EHP.json","view_paper":"https://pith.science/paper/PTZLBULT","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1806.08805&json=true","fetch_graph":"https://pith.science/api/pith-number/PTZLBULTCN4NJU64CF3XCM4EHP/graph.json","fetch_events":"https://pith.science/api/pith-number/PTZLBULTCN4NJU64CF3XCM4EHP/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/PTZLBULTCN4NJU64CF3XCM4EHP/action/timestamp_anchor","attest_storage":"https://pith.science/pith/PTZLBULTCN4NJU64CF3XCM4EHP/action/storage_attestation","attest_author":"https://pith.science/pith/PTZLBULTCN4NJU64CF3XCM4EHP/action/author_attestation","sign_citation":"https://pith.science/pith/PTZLBULTCN4NJU64CF3XCM4EHP/action/citation_signature","submit_replication":"https://pith.science/pith/PTZLBULTCN4NJU64CF3XCM4EHP/action/replication_record"}},"created_at":"2026-05-18T00:12:31.231380+00:00","updated_at":"2026-05-18T00:12:31.231380+00:00"}