{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2018:54VLWV52HPB6HYEECFWF55RNKC","short_pith_number":"pith:54VLWV52","schema_version":"1.0","canonical_sha256":"ef2abb57ba3bc3e3e084116c5ef62d50a959e640c4b9a806dd0f52c9fb92a3e3","source":{"kind":"arxiv","id":"1802.09766","version":6},"attestation_state":"computed","paper":{"title":"Learning Representations for Neural Network-Based Classification Using the Information Bottleneck Principle","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CV","cs.IT","math.IT"],"primary_cat":"cs.LG","authors_text":"Bernhard C. Geiger, Rana Ali Amjad","submitted_at":"2018-02-27T08:24:19Z","abstract_excerpt":"In this theory paper, we investigate training deep neural networks (DNNs) for classification via minimizing the information bottleneck (IB) functional. We show that the resulting optimization problem suffers from two severe issues: First, for deterministic DNNs, either the IB functional is infinite for almost all values of network parameters, making the optimization problem ill-posed, or it is piecewise constant, hence not admitting gradient-based optimization methods. Second, the invariance of the IB functional under bijections prevents it from capturing properties of the learned representati"},"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":"1802.09766","kind":"arxiv","version":6},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-02-27T08:24:19Z","cross_cats_sorted":["cs.CV","cs.IT","math.IT"],"title_canon_sha256":"8236673b508a504e21c2e12436589d1b0d3965a9b4360d108262cd0735345f54","abstract_canon_sha256":"af0443d134c8b5e4eb675f2a2a365d889d2db6d6425c3e4842343d578104257d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T01:25:20.089875Z","signature_b64":"6Jsu4++IoJ2c3EyJBI3q+0ho5kwtgd9r9G18nepvv3A0zukoBqU9JW91OKxOa4j3fYVuGpoXoWLJAsGuF+i/AQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ef2abb57ba3bc3e3e084116c5ef62d50a959e640c4b9a806dd0f52c9fb92a3e3","last_reissued_at":"2026-07-05T01:25:20.089385Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T01:25:20.089385Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Learning Representations for Neural Network-Based Classification Using the Information Bottleneck Principle","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CV","cs.IT","math.IT"],"primary_cat":"cs.LG","authors_text":"Bernhard C. Geiger, Rana Ali Amjad","submitted_at":"2018-02-27T08:24:19Z","abstract_excerpt":"In this theory paper, we investigate training deep neural networks (DNNs) for classification via minimizing the information bottleneck (IB) functional. We show that the resulting optimization problem suffers from two severe issues: First, for deterministic DNNs, either the IB functional is infinite for almost all values of network parameters, making the optimization problem ill-posed, or it is piecewise constant, hence not admitting gradient-based optimization methods. Second, the invariance of the IB functional under bijections prevents it from capturing properties of the learned representati"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1802.09766","kind":"arxiv","version":6},"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/1802.09766/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":"1802.09766","created_at":"2026-07-05T01:25:20.089446+00:00"},{"alias_kind":"arxiv_version","alias_value":"1802.09766v6","created_at":"2026-07-05T01:25:20.089446+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1802.09766","created_at":"2026-07-05T01:25:20.089446+00:00"},{"alias_kind":"pith_short_12","alias_value":"54VLWV52HPB6","created_at":"2026-07-05T01:25:20.089446+00:00"},{"alias_kind":"pith_short_16","alias_value":"54VLWV52HPB6HYEE","created_at":"2026-07-05T01:25:20.089446+00:00"},{"alias_kind":"pith_short_8","alias_value":"54VLWV52","created_at":"2026-07-05T01:25:20.089446+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.03636","citing_title":"Information Plane Analysis of Binary Neural Networks","ref_index":5,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/54VLWV52HPB6HYEECFWF55RNKC","json":"https://pith.science/pith/54VLWV52HPB6HYEECFWF55RNKC.json","graph_json":"https://pith.science/api/pith-number/54VLWV52HPB6HYEECFWF55RNKC/graph.json","events_json":"https://pith.science/api/pith-number/54VLWV52HPB6HYEECFWF55RNKC/events.json","paper":"https://pith.science/paper/54VLWV52"},"agent_actions":{"view_html":"https://pith.science/pith/54VLWV52HPB6HYEECFWF55RNKC","download_json":"https://pith.science/pith/54VLWV52HPB6HYEECFWF55RNKC.json","view_paper":"https://pith.science/paper/54VLWV52","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1802.09766&json=true","fetch_graph":"https://pith.science/api/pith-number/54VLWV52HPB6HYEECFWF55RNKC/graph.json","fetch_events":"https://pith.science/api/pith-number/54VLWV52HPB6HYEECFWF55RNKC/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/54VLWV52HPB6HYEECFWF55RNKC/action/timestamp_anchor","attest_storage":"https://pith.science/pith/54VLWV52HPB6HYEECFWF55RNKC/action/storage_attestation","attest_author":"https://pith.science/pith/54VLWV52HPB6HYEECFWF55RNKC/action/author_attestation","sign_citation":"https://pith.science/pith/54VLWV52HPB6HYEECFWF55RNKC/action/citation_signature","submit_replication":"https://pith.science/pith/54VLWV52HPB6HYEECFWF55RNKC/action/replication_record"}},"created_at":"2026-07-05T01:25:20.089446+00:00","updated_at":"2026-07-05T01:25:20.089446+00:00"}