{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2019:2PUF2JX6JBTNKS2FX6YN2Y5DHN","short_pith_number":"pith:2PUF2JX6","schema_version":"1.0","canonical_sha256":"d3e85d26fe4866d54b45bfb0dd63a33b6fa291e5802edf500180c52f8617cb52","source":{"kind":"arxiv","id":"1906.03671","version":2},"attestation_state":"computed","paper":{"title":"Deep Batch Active Learning by Diverse, Uncertain Gradient Lower Bounds","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Akshay Krishnamurthy, Alekh Agarwal, Chicheng Zhang, John Langford, Jordan T. Ash","submitted_at":"2019-06-09T16:52:09Z","abstract_excerpt":"We design a new algorithm for batch active learning with deep neural network models. Our algorithm, Batch Active learning by Diverse Gradient Embeddings (BADGE), samples groups of points that are disparate and high-magnitude when represented in a hallucinated gradient space, a strategy designed to incorporate both predictive uncertainty and sample diversity into every selected batch. Crucially, BADGE trades off between diversity and uncertainty without requiring any hand-tuned hyperparameters. We show that while other approaches sometimes succeed for particular batch sizes or architectures, BA"},"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":"1906.03671","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-06-09T16:52:09Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"c34e88d59dcea504899d919f0f68c10f9485459dd60570967d14e7ec2898cebe","abstract_canon_sha256":"a0e13ae50661decf32a835eaffa5161772f541a16bc4e5bd3ec0788b6d2cfd6c"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T00:43:01.445790Z","signature_b64":"rQP2pEee3dVznEJcR8N/lY3nnmXdtTJQ11ip3TopM0aQf0Kr14ZmnUwGZn+Gux4xUan1g2Bpq65ayFg+OjslCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d3e85d26fe4866d54b45bfb0dd63a33b6fa291e5802edf500180c52f8617cb52","last_reissued_at":"2026-07-05T00:43:01.445386Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T00:43:01.445386Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Deep Batch Active Learning by Diverse, Uncertain Gradient Lower Bounds","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Akshay Krishnamurthy, Alekh Agarwal, Chicheng Zhang, John Langford, Jordan T. Ash","submitted_at":"2019-06-09T16:52:09Z","abstract_excerpt":"We design a new algorithm for batch active learning with deep neural network models. Our algorithm, Batch Active learning by Diverse Gradient Embeddings (BADGE), samples groups of points that are disparate and high-magnitude when represented in a hallucinated gradient space, a strategy designed to incorporate both predictive uncertainty and sample diversity into every selected batch. Crucially, BADGE trades off between diversity and uncertainty without requiring any hand-tuned hyperparameters. We show that while other approaches sometimes succeed for particular batch sizes or architectures, BA"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1906.03671","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/1906.03671/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":"1906.03671","created_at":"2026-07-05T00:43:01.445440+00:00"},{"alias_kind":"arxiv_version","alias_value":"1906.03671v2","created_at":"2026-07-05T00:43:01.445440+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1906.03671","created_at":"2026-07-05T00:43:01.445440+00:00"},{"alias_kind":"pith_short_12","alias_value":"2PUF2JX6JBTN","created_at":"2026-07-05T00:43:01.445440+00:00"},{"alias_kind":"pith_short_16","alias_value":"2PUF2JX6JBTNKS2F","created_at":"2026-07-05T00:43:01.445440+00:00"},{"alias_kind":"pith_short_8","alias_value":"2PUF2JX6","created_at":"2026-07-05T00:43:01.445440+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":17,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.25301","citing_title":"Active Learning for Optimal Experimental Design in Machine Learning-Based Building Energy System Identification","ref_index":3,"is_internal_anchor":false},{"citing_arxiv_id":"2606.03821","citing_title":"Finding Needles in the Haystack: Transductive Active Labeling in Ecology","ref_index":49,"is_internal_anchor":false},{"citing_arxiv_id":"2606.03821","citing_title":"Finding Needles in the Haystack: Transductive Active Labeling in Ecology","ref_index":1,"is_internal_anchor":false},{"citing_arxiv_id":"2606.07630","citing_title":"Active Learning with Foundation Model Priors: Efficient Learning under Class Imbalance","ref_index":2,"is_internal_anchor":false},{"citing_arxiv_id":"2403.03208","citing_title":"Active Statistical Inference","ref_index":4,"is_internal_anchor":false},{"citing_arxiv_id":"2407.21772","citing_title":"ShieldGemma: Generative AI Content Moderation Based on Gemma","ref_index":2,"is_internal_anchor":false},{"citing_arxiv_id":"2510.03247","citing_title":"Towards Multimodal Active Learning: Efficient Learning with Limited Paired Data","ref_index":1,"is_internal_anchor":false},{"citing_arxiv_id":"2511.18107","citing_title":"Active Learning with Selective Time-Step Acquisition for PDEs","ref_index":1,"is_internal_anchor":false},{"citing_arxiv_id":"2605.14689","citing_title":"Are Candidate Models Really Needed for Active Learning?","ref_index":151,"is_internal_anchor":false},{"citing_arxiv_id":"2605.10784","citing_title":"MASS-DPO: Multi-negative Active Sample Selection for Direct Policy Optimization","ref_index":5,"is_internal_anchor":false},{"citing_arxiv_id":"2605.09858","citing_title":"Clip-level Uncertainty and Temporal-aware Active Learning for End-to-End Multi-Object Tracking","ref_index":15,"is_internal_anchor":false},{"citing_arxiv_id":"2604.25142","citing_title":"UnIte: Uncertainty-based Iterative Document Sampling for Domain Adaptation in Information Retrieval","ref_index":2,"is_internal_anchor":false},{"citing_arxiv_id":"2605.05590","citing_title":"Uncertainty-Guided Edge Learning for Deep Image Regression in Remote Sensing","ref_index":4,"is_internal_anchor":false},{"citing_arxiv_id":"2604.19027","citing_title":"Neural Operator Representation of Granular Micromechanics-based Failure Envelope","ref_index":1,"is_internal_anchor":false},{"citing_arxiv_id":"2604.11328","citing_title":"Select Smarter, Not More: Prompt-Aware Evaluation Scheduling with Submodular Guarantees","ref_index":37,"is_internal_anchor":false},{"citing_arxiv_id":"2604.12303","citing_title":"Labeled TrustSet Guided: Batch Active Learning with Reinforcement Learning","ref_index":1,"is_internal_anchor":false},{"citing_arxiv_id":"1909.08593","citing_title":"Fine-Tuning Language Models from Human Preferences","ref_index":1,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/2PUF2JX6JBTNKS2FX6YN2Y5DHN","json":"https://pith.science/pith/2PUF2JX6JBTNKS2FX6YN2Y5DHN.json","graph_json":"https://pith.science/api/pith-number/2PUF2JX6JBTNKS2FX6YN2Y5DHN/graph.json","events_json":"https://pith.science/api/pith-number/2PUF2JX6JBTNKS2FX6YN2Y5DHN/events.json","paper":"https://pith.science/paper/2PUF2JX6"},"agent_actions":{"view_html":"https://pith.science/pith/2PUF2JX6JBTNKS2FX6YN2Y5DHN","download_json":"https://pith.science/pith/2PUF2JX6JBTNKS2FX6YN2Y5DHN.json","view_paper":"https://pith.science/paper/2PUF2JX6","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1906.03671&json=true","fetch_graph":"https://pith.science/api/pith-number/2PUF2JX6JBTNKS2FX6YN2Y5DHN/graph.json","fetch_events":"https://pith.science/api/pith-number/2PUF2JX6JBTNKS2FX6YN2Y5DHN/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/2PUF2JX6JBTNKS2FX6YN2Y5DHN/action/timestamp_anchor","attest_storage":"https://pith.science/pith/2PUF2JX6JBTNKS2FX6YN2Y5DHN/action/storage_attestation","attest_author":"https://pith.science/pith/2PUF2JX6JBTNKS2FX6YN2Y5DHN/action/author_attestation","sign_citation":"https://pith.science/pith/2PUF2JX6JBTNKS2FX6YN2Y5DHN/action/citation_signature","submit_replication":"https://pith.science/pith/2PUF2JX6JBTNKS2FX6YN2Y5DHN/action/replication_record"}},"created_at":"2026-07-05T00:43:01.445440+00:00","updated_at":"2026-07-05T00:43:01.445440+00:00"}