{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:2YAJ7I5TXIVMKLLMGSBAYXJTKQ","short_pith_number":"pith:2YAJ7I5T","schema_version":"1.0","canonical_sha256":"d6009fa3b3ba2ac52d6c34820c5d33541ccf0bf82fbd96eff11b21339d219014","source":{"kind":"arxiv","id":"2511.18945","version":4},"attestation_state":"computed","paper":{"title":"MIST: Mutual Information Estimation Via Supervised Training","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["cs.IT","math.IT"],"primary_cat":"cs.LG","authors_text":"German Gritsai, Kyunghyun Cho, Maxime M\\'eloux, Maxime Peyrard, Megan Richards","submitted_at":"2025-11-24T09:55:28Z","abstract_excerpt":"We propose a fully data-driven approach to designing mutual information (MI) estimators. Since any MI estimator is a function of the observed sample from two random variables, we parameterize this function with a neural network (MIST) and train it end-to-end to predict MI values. Training is performed on a large meta-dataset of 625,000 synthetic joint distributions with known ground-truth MI. To handle variable sample sizes and dimensions, we employ a two-dimensional attention scheme ensuring permutation invariance across input samples. To quantify uncertainty, we optimize a quantile regressio"},"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":"2511.18945","kind":"arxiv","version":4},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.LG","submitted_at":"2025-11-24T09:55:28Z","cross_cats_sorted":["cs.IT","math.IT"],"title_canon_sha256":"024f02fc0edf67387e3dd37d64beb690593130edc005f9510608daa2428a0e1a","abstract_canon_sha256":"75c5f4e045ac8b99aea956fc0b3b6a0d3cf1d2c44837f1d79833d9271aa90463"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-08T01:05:04.621347Z","signature_b64":"Affd+k5IaJAJ7RO/Bp1rzMOjSNpBVdkGmYkJ5cY2+nH5T8J95j12KxQKDxPHgfAsliQK8drk6Xvc8qVlZ86ODg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d6009fa3b3ba2ac52d6c34820c5d33541ccf0bf82fbd96eff11b21339d219014","last_reissued_at":"2026-06-08T01:05:04.620281Z","signature_status":"signed_v1","first_computed_at":"2026-06-08T01:05:04.620281Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"MIST: Mutual Information Estimation Via Supervised Training","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["cs.IT","math.IT"],"primary_cat":"cs.LG","authors_text":"German Gritsai, Kyunghyun Cho, Maxime M\\'eloux, Maxime Peyrard, Megan Richards","submitted_at":"2025-11-24T09:55:28Z","abstract_excerpt":"We propose a fully data-driven approach to designing mutual information (MI) estimators. Since any MI estimator is a function of the observed sample from two random variables, we parameterize this function with a neural network (MIST) and train it end-to-end to predict MI values. Training is performed on a large meta-dataset of 625,000 synthetic joint distributions with known ground-truth MI. To handle variable sample sizes and dimensions, we employ a two-dimensional attention scheme ensuring permutation invariance across input samples. To quantify uncertainty, we optimize a quantile regressio"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2511.18945","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/2511.18945/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":"2511.18945","created_at":"2026-06-08T01:05:04.620430+00:00"},{"alias_kind":"arxiv_version","alias_value":"2511.18945v4","created_at":"2026-06-08T01:05:04.620430+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2511.18945","created_at":"2026-06-08T01:05:04.620430+00:00"},{"alias_kind":"pith_short_12","alias_value":"2YAJ7I5TXIVM","created_at":"2026-06-08T01:05:04.620430+00:00"},{"alias_kind":"pith_short_16","alias_value":"2YAJ7I5TXIVMKLLM","created_at":"2026-06-08T01:05:04.620430+00:00"},{"alias_kind":"pith_short_8","alias_value":"2YAJ7I5T","created_at":"2026-06-08T01:05:04.620430+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2604.20568","citing_title":"Amortized Vine Copulas for High-Dimensional Density and Information Estimation","ref_index":12,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/2YAJ7I5TXIVMKLLMGSBAYXJTKQ","json":"https://pith.science/pith/2YAJ7I5TXIVMKLLMGSBAYXJTKQ.json","graph_json":"https://pith.science/api/pith-number/2YAJ7I5TXIVMKLLMGSBAYXJTKQ/graph.json","events_json":"https://pith.science/api/pith-number/2YAJ7I5TXIVMKLLMGSBAYXJTKQ/events.json","paper":"https://pith.science/paper/2YAJ7I5T"},"agent_actions":{"view_html":"https://pith.science/pith/2YAJ7I5TXIVMKLLMGSBAYXJTKQ","download_json":"https://pith.science/pith/2YAJ7I5TXIVMKLLMGSBAYXJTKQ.json","view_paper":"https://pith.science/paper/2YAJ7I5T","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2511.18945&json=true","fetch_graph":"https://pith.science/api/pith-number/2YAJ7I5TXIVMKLLMGSBAYXJTKQ/graph.json","fetch_events":"https://pith.science/api/pith-number/2YAJ7I5TXIVMKLLMGSBAYXJTKQ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/2YAJ7I5TXIVMKLLMGSBAYXJTKQ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/2YAJ7I5TXIVMKLLMGSBAYXJTKQ/action/storage_attestation","attest_author":"https://pith.science/pith/2YAJ7I5TXIVMKLLMGSBAYXJTKQ/action/author_attestation","sign_citation":"https://pith.science/pith/2YAJ7I5TXIVMKLLMGSBAYXJTKQ/action/citation_signature","submit_replication":"https://pith.science/pith/2YAJ7I5TXIVMKLLMGSBAYXJTKQ/action/replication_record"}},"created_at":"2026-06-08T01:05:04.620430+00:00","updated_at":"2026-06-08T01:05:04.620430+00:00"}