{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:FO37LYPDNIYICXTVP6VHFYULSH","short_pith_number":"pith:FO37LYPD","schema_version":"1.0","canonical_sha256":"2bb7f5e1e36a30815e757faa72e28b91e3ac2925ed3ee879b9f55ce02a248644","source":{"kind":"arxiv","id":"2206.01816","version":2},"attestation_state":"computed","paper":{"title":"From $t$-SNE to UMAP with contrastive learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.HC"],"primary_cat":"cs.LG","authors_text":"(2) University of T\\\"ubingen), Dmitry Kobak (2) ((1) IWR at Heidelberg University, Fred A. Hamprecht (1), Jan Niklas B\\\"ohm (2), Sebastian Damrich (1)","submitted_at":"2022-06-03T20:50:54Z","abstract_excerpt":"Neighbor embedding methods $t$-SNE and UMAP are the de facto standard for visualizing high-dimensional datasets. Motivated from entirely different viewpoints, their loss functions appear to be unrelated. In practice, they yield strongly differing embeddings and can suggest conflicting interpretations of the same data. The fundamental reasons for this and, more generally, the exact relationship between $t$-SNE and UMAP have remained unclear. In this work, we uncover their conceptual connection via a new insight into contrastive learning methods. Noise-contrastive estimation can be used to optim"},"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":"2206.01816","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2022-06-03T20:50:54Z","cross_cats_sorted":["cs.HC"],"title_canon_sha256":"0aec91b3e04304fb70d415bb1e2180305d8cc8b161e02339748deae2051f64bc","abstract_canon_sha256":"9bb8f8e1ed1ce756563c961c44555fa2e0eb57ea95551d01be67451fb3516b97"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:27:21.668147Z","signature_b64":"p3SdMO7aWm0qt753JVo5lSDvjZvrnikCehHePzagUJd8SlargfoEOcHQqjVU3cRkjAS+cBKxvWciO4x4yX+xDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2bb7f5e1e36a30815e757faa72e28b91e3ac2925ed3ee879b9f55ce02a248644","last_reissued_at":"2026-07-05T08:27:21.667675Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:27:21.667675Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"From $t$-SNE to UMAP with contrastive learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.HC"],"primary_cat":"cs.LG","authors_text":"(2) University of T\\\"ubingen), Dmitry Kobak (2) ((1) IWR at Heidelberg University, Fred A. Hamprecht (1), Jan Niklas B\\\"ohm (2), Sebastian Damrich (1)","submitted_at":"2022-06-03T20:50:54Z","abstract_excerpt":"Neighbor embedding methods $t$-SNE and UMAP are the de facto standard for visualizing high-dimensional datasets. Motivated from entirely different viewpoints, their loss functions appear to be unrelated. In practice, they yield strongly differing embeddings and can suggest conflicting interpretations of the same data. The fundamental reasons for this and, more generally, the exact relationship between $t$-SNE and UMAP have remained unclear. In this work, we uncover their conceptual connection via a new insight into contrastive learning methods. Noise-contrastive estimation can be used to optim"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2206.01816","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/2206.01816/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"},"aliases":[{"alias_kind":"arxiv","alias_value":"2206.01816","created_at":"2026-07-05T08:27:21.667734+00:00"},{"alias_kind":"arxiv_version","alias_value":"2206.01816v2","created_at":"2026-07-05T08:27:21.667734+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2206.01816","created_at":"2026-07-05T08:27:21.667734+00:00"},{"alias_kind":"pith_short_12","alias_value":"FO37LYPDNIYI","created_at":"2026-07-05T08:27:21.667734+00:00"},{"alias_kind":"pith_short_16","alias_value":"FO37LYPDNIYICXTV","created_at":"2026-07-05T08:27:21.667734+00:00"},{"alias_kind":"pith_short_8","alias_value":"FO37LYPD","created_at":"2026-07-05T08:27:21.667734+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2510.18935","citing_title":"Feature Extraction in the Remote Sensing Data Value Chain: A Systematic Review of Methods and Applications","ref_index":225,"is_internal_anchor":false},{"citing_arxiv_id":"2604.02535","citing_title":"A Spectral Framework for Multi-Scale Nonlinear Dimensionality Reduction","ref_index":16,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/FO37LYPDNIYICXTVP6VHFYULSH","json":"https://pith.science/pith/FO37LYPDNIYICXTVP6VHFYULSH.json","graph_json":"https://pith.science/api/pith-number/FO37LYPDNIYICXTVP6VHFYULSH/graph.json","events_json":"https://pith.science/api/pith-number/FO37LYPDNIYICXTVP6VHFYULSH/events.json","paper":"https://pith.science/paper/FO37LYPD"},"agent_actions":{"view_html":"https://pith.science/pith/FO37LYPDNIYICXTVP6VHFYULSH","download_json":"https://pith.science/pith/FO37LYPDNIYICXTVP6VHFYULSH.json","view_paper":"https://pith.science/paper/FO37LYPD","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2206.01816&json=true","fetch_graph":"https://pith.science/api/pith-number/FO37LYPDNIYICXTVP6VHFYULSH/graph.json","fetch_events":"https://pith.science/api/pith-number/FO37LYPDNIYICXTVP6VHFYULSH/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/FO37LYPDNIYICXTVP6VHFYULSH/action/timestamp_anchor","attest_storage":"https://pith.science/pith/FO37LYPDNIYICXTVP6VHFYULSH/action/storage_attestation","attest_author":"https://pith.science/pith/FO37LYPDNIYICXTVP6VHFYULSH/action/author_attestation","sign_citation":"https://pith.science/pith/FO37LYPDNIYICXTVP6VHFYULSH/action/citation_signature","submit_replication":"https://pith.science/pith/FO37LYPDNIYICXTVP6VHFYULSH/action/replication_record"}},"created_at":"2026-07-05T08:27:21.667734+00:00","updated_at":"2026-07-05T08:27:21.667734+00:00"}