{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:Z2G2DTY57VCO7S3EUYGZLQ5BNT","short_pith_number":"pith:Z2G2DTY5","schema_version":"1.0","canonical_sha256":"ce8da1cf1dfd44efcb64a60d95c3a16cf6ab7922655c40bffefd98c017413dba","source":{"kind":"arxiv","id":"2106.05238","version":4},"attestation_state":"computed","paper":{"title":"I Don't Need u: Identifiable Non-Linear ICA Without Side Information","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CV","eess.SP","stat.ML"],"primary_cat":"cs.LG","authors_text":"Brooks Paige, Matthew Willetts","submitted_at":"2021-06-09T17:22:08Z","abstract_excerpt":"In this paper, we investigate the algorithmic stability of unsupervised representation learning with deep generative models, as a function of repeated re-training on the same input data. Algorithms for learning low dimensional linear representations -- for example principal components analysis (PCA), or linear independent components analysis (ICA) -- come with guarantees that they will always reveal the same latent representations (perhaps up to an arbitrary rotation or permutation). Unfortunately, for non-linear representation learning, such as in a variational auto-encoder (VAE) model traine"},"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":"2106.05238","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2021-06-09T17:22:08Z","cross_cats_sorted":["cs.CV","eess.SP","stat.ML"],"title_canon_sha256":"b7ed1b09bc2d94a8a0aa79843915100c4fbeda94f308fe01f03744babc0a98b7","abstract_canon_sha256":"190047659b40d638e8281e7112d0f745d80665c5327d21412201fb3e1ac572a3"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:37:01.861506Z","signature_b64":"QNYWAEd5vr5kn1KESl592A/e/cxKduGcNf1ENiHWv6UJVPiEiaKc/NAAdxEnjRvKf3v8qE1tjvJzu3Z+OTApAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ce8da1cf1dfd44efcb64a60d95c3a16cf6ab7922655c40bffefd98c017413dba","last_reissued_at":"2026-07-05T04:37:01.861060Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:37:01.861060Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"I Don't Need u: Identifiable Non-Linear ICA Without Side Information","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CV","eess.SP","stat.ML"],"primary_cat":"cs.LG","authors_text":"Brooks Paige, Matthew Willetts","submitted_at":"2021-06-09T17:22:08Z","abstract_excerpt":"In this paper, we investigate the algorithmic stability of unsupervised representation learning with deep generative models, as a function of repeated re-training on the same input data. Algorithms for learning low dimensional linear representations -- for example principal components analysis (PCA), or linear independent components analysis (ICA) -- come with guarantees that they will always reveal the same latent representations (perhaps up to an arbitrary rotation or permutation). Unfortunately, for non-linear representation learning, such as in a variational auto-encoder (VAE) model traine"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2106.05238","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":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2106.05238/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":"2106.05238","created_at":"2026-07-05T04:37:01.861118+00:00"},{"alias_kind":"arxiv_version","alias_value":"2106.05238v4","created_at":"2026-07-05T04:37:01.861118+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2106.05238","created_at":"2026-07-05T04:37:01.861118+00:00"},{"alias_kind":"pith_short_12","alias_value":"Z2G2DTY57VCO","created_at":"2026-07-05T04:37:01.861118+00:00"},{"alias_kind":"pith_short_16","alias_value":"Z2G2DTY57VCO7S3E","created_at":"2026-07-05T04:37:01.861118+00:00"},{"alias_kind":"pith_short_8","alias_value":"Z2G2DTY5","created_at":"2026-07-05T04:37:01.861118+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2603.23547","citing_title":"PDGMM-VAE: A Variational Autoencoder with Adaptive Per-Dimension Gaussian Mixture Model Priors for Nonlinear ICA","ref_index":11,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/Z2G2DTY57VCO7S3EUYGZLQ5BNT","json":"https://pith.science/pith/Z2G2DTY57VCO7S3EUYGZLQ5BNT.json","graph_json":"https://pith.science/api/pith-number/Z2G2DTY57VCO7S3EUYGZLQ5BNT/graph.json","events_json":"https://pith.science/api/pith-number/Z2G2DTY57VCO7S3EUYGZLQ5BNT/events.json","paper":"https://pith.science/paper/Z2G2DTY5"},"agent_actions":{"view_html":"https://pith.science/pith/Z2G2DTY57VCO7S3EUYGZLQ5BNT","download_json":"https://pith.science/pith/Z2G2DTY57VCO7S3EUYGZLQ5BNT.json","view_paper":"https://pith.science/paper/Z2G2DTY5","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2106.05238&json=true","fetch_graph":"https://pith.science/api/pith-number/Z2G2DTY57VCO7S3EUYGZLQ5BNT/graph.json","fetch_events":"https://pith.science/api/pith-number/Z2G2DTY57VCO7S3EUYGZLQ5BNT/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/Z2G2DTY57VCO7S3EUYGZLQ5BNT/action/timestamp_anchor","attest_storage":"https://pith.science/pith/Z2G2DTY57VCO7S3EUYGZLQ5BNT/action/storage_attestation","attest_author":"https://pith.science/pith/Z2G2DTY57VCO7S3EUYGZLQ5BNT/action/author_attestation","sign_citation":"https://pith.science/pith/Z2G2DTY57VCO7S3EUYGZLQ5BNT/action/citation_signature","submit_replication":"https://pith.science/pith/Z2G2DTY57VCO7S3EUYGZLQ5BNT/action/replication_record"}},"created_at":"2026-07-05T04:37:01.861118+00:00","updated_at":"2026-07-05T04:37:01.861118+00:00"}