{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:B6J7FXFLIMJNS3Y5P52M4AGECK","short_pith_number":"pith:B6J7FXFL","schema_version":"1.0","canonical_sha256":"0f93f2dcab4312d96f1d7f74ce00c412b0851a49998ac74d63dfedae284da3cf","source":{"kind":"arxiv","id":"2301.09490","version":1},"attestation_state":"computed","paper":{"title":"Speeding Up BatchBALD: A k-BALD Family of Approximations for Active Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Andreas Kirsch","submitted_at":"2023-01-23T15:38:58Z","abstract_excerpt":"Active learning is a powerful method for training machine learning models with limited labeled data. One commonly used technique for active learning is BatchBALD, which uses Bayesian neural networks to find the most informative points to label in a pool set. However, BatchBALD can be very slow to compute, especially for larger datasets. In this paper, we propose a new approximation, k-BALD, which uses k-wise mutual information terms to approximate BatchBALD, making it much less expensive to compute. Results on the MNIST dataset show that k-BALD is significantly faster than BatchBALD while main"},"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":"2301.09490","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2023-01-23T15:38:58Z","cross_cats_sorted":[],"title_canon_sha256":"38992b47ce5084e3ec2ce642320ec1b451cccc6b5484f40ce4fb59dd07626d55","abstract_canon_sha256":"dd48fe9110d8819eeeabfc0c0ef4a68074897d5deb315d23b825e392197d2e99"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:35:01.369952Z","signature_b64":"qU7itc9CJkB6FYc5FAcTBhf1Y1PaFyE3R82Tym1827xD0KeMHY2uVEd932H+i+xmfxttNTTem1ryBiEBqZsfAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0f93f2dcab4312d96f1d7f74ce00c412b0851a49998ac74d63dfedae284da3cf","last_reissued_at":"2026-07-05T05:35:01.369403Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:35:01.369403Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Speeding Up BatchBALD: A k-BALD Family of Approximations for Active Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Andreas Kirsch","submitted_at":"2023-01-23T15:38:58Z","abstract_excerpt":"Active learning is a powerful method for training machine learning models with limited labeled data. One commonly used technique for active learning is BatchBALD, which uses Bayesian neural networks to find the most informative points to label in a pool set. However, BatchBALD can be very slow to compute, especially for larger datasets. In this paper, we propose a new approximation, k-BALD, which uses k-wise mutual information terms to approximate BatchBALD, making it much less expensive to compute. Results on the MNIST dataset show that k-BALD is significantly faster than BatchBALD while main"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2301.09490","kind":"arxiv","version":1},"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/2301.09490/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":"2301.09490","created_at":"2026-07-05T05:35:01.369464+00:00"},{"alias_kind":"arxiv_version","alias_value":"2301.09490v1","created_at":"2026-07-05T05:35:01.369464+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2301.09490","created_at":"2026-07-05T05:35:01.369464+00:00"},{"alias_kind":"pith_short_12","alias_value":"B6J7FXFLIMJN","created_at":"2026-07-05T05:35:01.369464+00:00"},{"alias_kind":"pith_short_16","alias_value":"B6J7FXFLIMJNS3Y5","created_at":"2026-07-05T05:35:01.369464+00:00"},{"alias_kind":"pith_short_8","alias_value":"B6J7FXFL","created_at":"2026-07-05T05:35:01.369464+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2507.04929","citing_title":"ConBatch-BAL: Batch Bayesian Active Learning under Budget Constraints","ref_index":21,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/B6J7FXFLIMJNS3Y5P52M4AGECK","json":"https://pith.science/pith/B6J7FXFLIMJNS3Y5P52M4AGECK.json","graph_json":"https://pith.science/api/pith-number/B6J7FXFLIMJNS3Y5P52M4AGECK/graph.json","events_json":"https://pith.science/api/pith-number/B6J7FXFLIMJNS3Y5P52M4AGECK/events.json","paper":"https://pith.science/paper/B6J7FXFL"},"agent_actions":{"view_html":"https://pith.science/pith/B6J7FXFLIMJNS3Y5P52M4AGECK","download_json":"https://pith.science/pith/B6J7FXFLIMJNS3Y5P52M4AGECK.json","view_paper":"https://pith.science/paper/B6J7FXFL","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2301.09490&json=true","fetch_graph":"https://pith.science/api/pith-number/B6J7FXFLIMJNS3Y5P52M4AGECK/graph.json","fetch_events":"https://pith.science/api/pith-number/B6J7FXFLIMJNS3Y5P52M4AGECK/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/B6J7FXFLIMJNS3Y5P52M4AGECK/action/timestamp_anchor","attest_storage":"https://pith.science/pith/B6J7FXFLIMJNS3Y5P52M4AGECK/action/storage_attestation","attest_author":"https://pith.science/pith/B6J7FXFLIMJNS3Y5P52M4AGECK/action/author_attestation","sign_citation":"https://pith.science/pith/B6J7FXFLIMJNS3Y5P52M4AGECK/action/citation_signature","submit_replication":"https://pith.science/pith/B6J7FXFLIMJNS3Y5P52M4AGECK/action/replication_record"}},"created_at":"2026-07-05T05:35:01.369464+00:00","updated_at":"2026-07-05T05:35:01.369464+00:00"}