{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2019:7JFORSTFOO2PS4MN47WTGORPWN","short_pith_number":"pith:7JFORSTF","schema_version":"1.0","canonical_sha256":"fa4ae8ca6573b4f9718de7ed333a2fb367c17bb67a1b1748f266c24e71c78b64","source":{"kind":"arxiv","id":"1907.13625","version":2},"attestation_state":"computed","paper":{"title":"On Mutual Information Maximization for Representation Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Josip Djolonga, Mario Lucic, Michael Tschannen, Paul K. Rubenstein, Sylvain Gelly","submitted_at":"2019-07-31T17:50:51Z","abstract_excerpt":"Many recent methods for unsupervised or self-supervised representation learning train feature extractors by maximizing an estimate of the mutual information (MI) between different views of the data. This comes with several immediate problems: For example, MI is notoriously hard to estimate, and using it as an objective for representation learning may lead to highly entangled representations due to its invariance under arbitrary invertible transformations. Nevertheless, these methods have been repeatedly shown to excel in practice. In this paper we argue, and provide empirical evidence, that th"},"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":"1907.13625","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-07-31T17:50:51Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"9973f83913b68de7e679666c68165e1af1eaabf2ba94a33b2a614fe79cae640b","abstract_canon_sha256":"59f20ff744b94fe1cb3e4e1bd3ad5b136eddd0d94282f3668cc7e625c2e8c069"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T00:35:10.158892Z","signature_b64":"W92JbLIgWwuZbf7wjzbJ8+OLkRZyv0zN8ic2TCwSuW3Za+i/k2m1ez5lTlZ21QfVt0MH4UqRqMoG91tY9CuMAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"fa4ae8ca6573b4f9718de7ed333a2fb367c17bb67a1b1748f266c24e71c78b64","last_reissued_at":"2026-07-05T00:35:10.158406Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T00:35:10.158406Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"On Mutual Information Maximization for Representation Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Josip Djolonga, Mario Lucic, Michael Tschannen, Paul K. Rubenstein, Sylvain Gelly","submitted_at":"2019-07-31T17:50:51Z","abstract_excerpt":"Many recent methods for unsupervised or self-supervised representation learning train feature extractors by maximizing an estimate of the mutual information (MI) between different views of the data. This comes with several immediate problems: For example, MI is notoriously hard to estimate, and using it as an objective for representation learning may lead to highly entangled representations due to its invariance under arbitrary invertible transformations. Nevertheless, these methods have been repeatedly shown to excel in practice. In this paper we argue, and provide empirical evidence, that th"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1907.13625","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/1907.13625/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":"1907.13625","created_at":"2026-07-05T00:35:10.158463+00:00"},{"alias_kind":"arxiv_version","alias_value":"1907.13625v2","created_at":"2026-07-05T00:35:10.158463+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1907.13625","created_at":"2026-07-05T00:35:10.158463+00:00"},{"alias_kind":"pith_short_12","alias_value":"7JFORSTFOO2P","created_at":"2026-07-05T00:35:10.158463+00:00"},{"alias_kind":"pith_short_16","alias_value":"7JFORSTFOO2PS4MN","created_at":"2026-07-05T00:35:10.158463+00:00"},{"alias_kind":"pith_short_8","alias_value":"7JFORSTF","created_at":"2026-07-05T00:35:10.158463+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":6,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.00241","citing_title":"InfoAtlas: A Foundation Model for Zero-Shot Statistical Dependence Estimate","ref_index":100,"is_internal_anchor":false},{"citing_arxiv_id":"1910.04867","citing_title":"A Large-scale Study of Representation Learning with the Visual Task Adaptation Benchmark","ref_index":17,"is_internal_anchor":false},{"citing_arxiv_id":"2605.12536","citing_title":"Information as Maximum-Caliber Deviation: A bridge between Integrated Information Theory and the Free Energy Principle","ref_index":157,"is_internal_anchor":false},{"citing_arxiv_id":"2002.05709","citing_title":"A Simple Framework for Contrastive Learning of Visual Representations","ref_index":53,"is_internal_anchor":false},{"citing_arxiv_id":"2404.08471","citing_title":"Revisiting Feature Prediction for Learning Visual Representations from Video","ref_index":30,"is_internal_anchor":false},{"citing_arxiv_id":"2604.08936","citing_title":"M-IDoL: Information Decomposition for Modality-Specific and Diverse Representation Learning in Medical Foundation Model","ref_index":16,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/7JFORSTFOO2PS4MN47WTGORPWN","json":"https://pith.science/pith/7JFORSTFOO2PS4MN47WTGORPWN.json","graph_json":"https://pith.science/api/pith-number/7JFORSTFOO2PS4MN47WTGORPWN/graph.json","events_json":"https://pith.science/api/pith-number/7JFORSTFOO2PS4MN47WTGORPWN/events.json","paper":"https://pith.science/paper/7JFORSTF"},"agent_actions":{"view_html":"https://pith.science/pith/7JFORSTFOO2PS4MN47WTGORPWN","download_json":"https://pith.science/pith/7JFORSTFOO2PS4MN47WTGORPWN.json","view_paper":"https://pith.science/paper/7JFORSTF","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1907.13625&json=true","fetch_graph":"https://pith.science/api/pith-number/7JFORSTFOO2PS4MN47WTGORPWN/graph.json","fetch_events":"https://pith.science/api/pith-number/7JFORSTFOO2PS4MN47WTGORPWN/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/7JFORSTFOO2PS4MN47WTGORPWN/action/timestamp_anchor","attest_storage":"https://pith.science/pith/7JFORSTFOO2PS4MN47WTGORPWN/action/storage_attestation","attest_author":"https://pith.science/pith/7JFORSTFOO2PS4MN47WTGORPWN/action/author_attestation","sign_citation":"https://pith.science/pith/7JFORSTFOO2PS4MN47WTGORPWN/action/citation_signature","submit_replication":"https://pith.science/pith/7JFORSTFOO2PS4MN47WTGORPWN/action/replication_record"}},"created_at":"2026-07-05T00:35:10.158463+00:00","updated_at":"2026-07-05T00:35:10.158463+00:00"}