{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2018:XKUKNTGPUI3CIDPTWRQ3YKXHQY","short_pith_number":"pith:XKUKNTGP","schema_version":"1.0","canonical_sha256":"baa8a6cccfa236240df3b461bc2ae78637936748ed43880d8addf9dacc9f760b","source":{"kind":"arxiv","id":"1810.00821","version":4},"attestation_state":"computed","paper":{"title":"Variational Discriminator Bottleneck: Improving Imitation Learning, Inverse RL, and GANs by Constraining Information Flow","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Angjoo Kanazawa, Pieter Abbeel, Sam Toyer, Sergey Levine, Xue Bin Peng","submitted_at":"2018-10-01T17:02:24Z","abstract_excerpt":"Adversarial learning methods have been proposed for a wide range of applications, but the training of adversarial models can be notoriously unstable. Effectively balancing the performance of the generator and discriminator is critical, since a discriminator that achieves very high accuracy will produce relatively uninformative gradients. In this work, we propose a simple and general technique to constrain information flow in the discriminator by means of an information bottleneck. By enforcing a constraint on the mutual information between the observations and the discriminator's internal repr"},"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":"1810.00821","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-10-01T17:02:24Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"d9173a5b7808edd69b1d6d7068887d8892e4c5216aec31331d979126eb83f616","abstract_canon_sha256":"4a5fb0e3c18dd368df686357ecbc4fa203fb5a1779d7af864a86f64a669596d2"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T01:29:26.398652Z","signature_b64":"cSBu1tdE14PsRkYGwL5ApL1vJzHOQhT8PreRbIA74oQaAjuCB+x0oVwlBZZaQST7m6se65aUG+OkLlE048nQAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"baa8a6cccfa236240df3b461bc2ae78637936748ed43880d8addf9dacc9f760b","last_reissued_at":"2026-07-05T01:29:26.398146Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T01:29:26.398146Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Variational Discriminator Bottleneck: Improving Imitation Learning, Inverse RL, and GANs by Constraining Information Flow","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Angjoo Kanazawa, Pieter Abbeel, Sam Toyer, Sergey Levine, Xue Bin Peng","submitted_at":"2018-10-01T17:02:24Z","abstract_excerpt":"Adversarial learning methods have been proposed for a wide range of applications, but the training of adversarial models can be notoriously unstable. Effectively balancing the performance of the generator and discriminator is critical, since a discriminator that achieves very high accuracy will produce relatively uninformative gradients. In this work, we propose a simple and general technique to constrain information flow in the discriminator by means of an information bottleneck. By enforcing a constraint on the mutual information between the observations and the discriminator's internal repr"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1810.00821","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/1810.00821/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":"1810.00821","created_at":"2026-07-05T01:29:26.398204+00:00"},{"alias_kind":"arxiv_version","alias_value":"1810.00821v4","created_at":"2026-07-05T01:29:26.398204+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1810.00821","created_at":"2026-07-05T01:29:26.398204+00:00"},{"alias_kind":"pith_short_12","alias_value":"XKUKNTGPUI3C","created_at":"2026-07-05T01:29:26.398204+00:00"},{"alias_kind":"pith_short_16","alias_value":"XKUKNTGPUI3CIDPT","created_at":"2026-07-05T01:29:26.398204+00:00"},{"alias_kind":"pith_short_8","alias_value":"XKUKNTGP","created_at":"2026-07-05T01:29:26.398204+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2508.07652","citing_title":"Estimating classical mutual information between quantum subsystems with neural networks","ref_index":17,"is_internal_anchor":false},{"citing_arxiv_id":"2604.18460","citing_title":"Learning Invariant Modality Representation for Robust Multimodal Learning from a Causal Inference Perspective","ref_index":119,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/XKUKNTGPUI3CIDPTWRQ3YKXHQY","json":"https://pith.science/pith/XKUKNTGPUI3CIDPTWRQ3YKXHQY.json","graph_json":"https://pith.science/api/pith-number/XKUKNTGPUI3CIDPTWRQ3YKXHQY/graph.json","events_json":"https://pith.science/api/pith-number/XKUKNTGPUI3CIDPTWRQ3YKXHQY/events.json","paper":"https://pith.science/paper/XKUKNTGP"},"agent_actions":{"view_html":"https://pith.science/pith/XKUKNTGPUI3CIDPTWRQ3YKXHQY","download_json":"https://pith.science/pith/XKUKNTGPUI3CIDPTWRQ3YKXHQY.json","view_paper":"https://pith.science/paper/XKUKNTGP","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1810.00821&json=true","fetch_graph":"https://pith.science/api/pith-number/XKUKNTGPUI3CIDPTWRQ3YKXHQY/graph.json","fetch_events":"https://pith.science/api/pith-number/XKUKNTGPUI3CIDPTWRQ3YKXHQY/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/XKUKNTGPUI3CIDPTWRQ3YKXHQY/action/timestamp_anchor","attest_storage":"https://pith.science/pith/XKUKNTGPUI3CIDPTWRQ3YKXHQY/action/storage_attestation","attest_author":"https://pith.science/pith/XKUKNTGPUI3CIDPTWRQ3YKXHQY/action/author_attestation","sign_citation":"https://pith.science/pith/XKUKNTGPUI3CIDPTWRQ3YKXHQY/action/citation_signature","submit_replication":"https://pith.science/pith/XKUKNTGPUI3CIDPTWRQ3YKXHQY/action/replication_record"}},"created_at":"2026-07-05T01:29:26.398204+00:00","updated_at":"2026-07-05T01:29:26.398204+00:00"}