{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2016:G6IQBMPAFNS7AIYPVK3ZQRWJ6W","short_pith_number":"pith:G6IQBMPA","schema_version":"1.0","canonical_sha256":"379100b1e02b65f0230faab79846c9f5a56d2e6c8020b65fbdb82cb54c48dde7","source":{"kind":"arxiv","id":"1610.06545","version":4},"attestation_state":"computed","paper":{"title":"Revisiting Classifier Two-Sample Tests","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"stat.ML","authors_text":"David Lopez-Paz, Maxime Oquab","submitted_at":"2016-10-20T19:16:10Z","abstract_excerpt":"The goal of two-sample tests is to assess whether two samples, $S_P \\sim P^n$ and $S_Q \\sim Q^m$, are drawn from the same distribution. Perhaps intriguingly, one relatively unexplored method to build two-sample tests is the use of binary classifiers. In particular, construct a dataset by pairing the $n$ examples in $S_P$ with a positive label, and by pairing the $m$ examples in $S_Q$ with a negative label. If the null hypothesis \"$P = Q$\" is true, then the classification accuracy of a binary classifier on a held-out subset of this dataset should remain near chance-level. As we will show, such "},"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":"1610.06545","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"stat.ML","submitted_at":"2016-10-20T19:16:10Z","cross_cats_sorted":[],"title_canon_sha256":"6de7647f51da874ecdb82a243330b0ef6786aa2db19ff09d150c49d5d4b9a522","abstract_canon_sha256":"d82a8aedc1a9687449c3272b10f3516e25ddb23cf6d73fa42eafc5dcb20a9978"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T00:21:29.140401Z","signature_b64":"kDQYoh75c/o3fqMMSSkNB1/ZOJjb/wfPxWhXl/ULPmX96BhWrPF6H9rl2+sUqFwAdZpcXNMvhvo+roRjTVRhBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"379100b1e02b65f0230faab79846c9f5a56d2e6c8020b65fbdb82cb54c48dde7","last_reissued_at":"2026-05-18T00:21:29.139730Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T00:21:29.139730Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Revisiting Classifier Two-Sample Tests","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"stat.ML","authors_text":"David Lopez-Paz, Maxime Oquab","submitted_at":"2016-10-20T19:16:10Z","abstract_excerpt":"The goal of two-sample tests is to assess whether two samples, $S_P \\sim P^n$ and $S_Q \\sim Q^m$, are drawn from the same distribution. Perhaps intriguingly, one relatively unexplored method to build two-sample tests is the use of binary classifiers. In particular, construct a dataset by pairing the $n$ examples in $S_P$ with a positive label, and by pairing the $m$ examples in $S_Q$ with a negative label. If the null hypothesis \"$P = Q$\" is true, then the classification accuracy of a binary classifier on a held-out subset of this dataset should remain near chance-level. As we will show, such "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1610.06545","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"},"aliases":[{"alias_kind":"arxiv","alias_value":"1610.06545","created_at":"2026-05-18T00:21:29.139822+00:00"},{"alias_kind":"arxiv_version","alias_value":"1610.06545v4","created_at":"2026-05-18T00:21:29.139822+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1610.06545","created_at":"2026-05-18T00:21:29.139822+00:00"},{"alias_kind":"pith_short_12","alias_value":"G6IQBMPAFNS7","created_at":"2026-05-18T12:30:15.759754+00:00"},{"alias_kind":"pith_short_16","alias_value":"G6IQBMPAFNS7AIYP","created_at":"2026-05-18T12:30:15.759754+00:00"},{"alias_kind":"pith_short_8","alias_value":"G6IQBMPA","created_at":"2026-05-18T12:30:15.759754+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":15,"internal_anchor_count":8,"sample":[{"citing_arxiv_id":"2605.08179","citing_title":"Neural Posterior Estimation of Terrain Parameters from Radar Sounder Data","ref_index":36,"is_internal_anchor":true},{"citing_arxiv_id":"2605.27499","citing_title":"GenSBI: Generative Methods for Simulation-Based Inference in JAX","ref_index":80,"is_internal_anchor":true},{"citing_arxiv_id":"2601.22367","citing_title":"Amortized Simulation-Based Inference in Generalized Bayes via Neural Posterior Estimation","ref_index":9,"is_internal_anchor":true},{"citing_arxiv_id":"1907.08127","citing_title":"A discriminative approach for finding and characterizing positivity violations using decision trees","ref_index":9,"is_internal_anchor":true},{"citing_arxiv_id":"2412.00404","citing_title":"Hard-Label Black-Box Attacks on 3D Point Clouds","ref_index":91,"is_internal_anchor":true},{"citing_arxiv_id":"2605.17620","citing_title":"SynVA: A Modular Toolkit for Vessel Generation and Aneurysm Editing","ref_index":49,"is_internal_anchor":true},{"citing_arxiv_id":"2601.11727","citing_title":"Asymptotically Optimal Tests for One- and Two-Sample Problems","ref_index":7,"is_internal_anchor":true},{"citing_arxiv_id":"1801.01401","citing_title":"Demystifying MMD GANs","ref_index":33,"is_internal_anchor":true},{"citing_arxiv_id":"2605.10719","citing_title":"Machine Learning Techniques for Astrophysics and Cosmology: Simulation-Based Inference","ref_index":37,"is_internal_anchor":false},{"citing_arxiv_id":"2605.08179","citing_title":"Neural Posterior Estimation of Terrain Parameters from Radar Sounder Data","ref_index":36,"is_internal_anchor":false},{"citing_arxiv_id":"2605.02222","citing_title":"Generative Modeling with Orbit-Space Particle Flow Matching","ref_index":126,"is_internal_anchor":false},{"citing_arxiv_id":"2605.05520","citing_title":"Bayesian Rain Field Reconstruction using Commercial Microwave Links and Diffusion Model Priors","ref_index":145,"is_internal_anchor":false},{"citing_arxiv_id":"2604.13157","citing_title":"Proton Structure from Neural Simulation-Based Inference at the LHC","ref_index":132,"is_internal_anchor":false},{"citing_arxiv_id":"2604.10967","citing_title":"Learning to Test: Physics-Informed Representation for Dynamical Instability Detection","ref_index":48,"is_internal_anchor":false},{"citing_arxiv_id":"2605.07765","citing_title":"Pre-trained Tabular Foundation Models as Versatile Summary Networks for Neural Posterior Estimation","ref_index":14,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/G6IQBMPAFNS7AIYPVK3ZQRWJ6W","json":"https://pith.science/pith/G6IQBMPAFNS7AIYPVK3ZQRWJ6W.json","graph_json":"https://pith.science/api/pith-number/G6IQBMPAFNS7AIYPVK3ZQRWJ6W/graph.json","events_json":"https://pith.science/api/pith-number/G6IQBMPAFNS7AIYPVK3ZQRWJ6W/events.json","paper":"https://pith.science/paper/G6IQBMPA"},"agent_actions":{"view_html":"https://pith.science/pith/G6IQBMPAFNS7AIYPVK3ZQRWJ6W","download_json":"https://pith.science/pith/G6IQBMPAFNS7AIYPVK3ZQRWJ6W.json","view_paper":"https://pith.science/paper/G6IQBMPA","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1610.06545&json=true","fetch_graph":"https://pith.science/api/pith-number/G6IQBMPAFNS7AIYPVK3ZQRWJ6W/graph.json","fetch_events":"https://pith.science/api/pith-number/G6IQBMPAFNS7AIYPVK3ZQRWJ6W/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/G6IQBMPAFNS7AIYPVK3ZQRWJ6W/action/timestamp_anchor","attest_storage":"https://pith.science/pith/G6IQBMPAFNS7AIYPVK3ZQRWJ6W/action/storage_attestation","attest_author":"https://pith.science/pith/G6IQBMPAFNS7AIYPVK3ZQRWJ6W/action/author_attestation","sign_citation":"https://pith.science/pith/G6IQBMPAFNS7AIYPVK3ZQRWJ6W/action/citation_signature","submit_replication":"https://pith.science/pith/G6IQBMPAFNS7AIYPVK3ZQRWJ6W/action/replication_record"}},"created_at":"2026-05-18T00:21:29.139822+00:00","updated_at":"2026-05-18T00:21:29.139822+00:00"}