{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:I6JNOZXSQPD4CCD3NO3PD7CHTP","short_pith_number":"pith:I6JNOZXS","schema_version":"1.0","canonical_sha256":"4792d766f283c7c1087b6bb6f1fc479bf9652c1e48a56d3a7a27938212f4eeea","source":{"kind":"arxiv","id":"2305.06408","version":2},"attestation_state":"computed","paper":{"title":"Accelerating Batch Active Learning Using Continual Learning Techniques","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Arnav Das, Gantavya Bhatt, Jeff Bilmes, Megh Bhalerao, Rui Yang, Vianne Gao","submitted_at":"2023-05-10T18:34:05Z","abstract_excerpt":"A major problem with Active Learning (AL) is high training costs since models are typically retrained from scratch after every query round. We start by demonstrating that standard AL on neural networks with warm starting fails, both to accelerate training and to avoid catastrophic forgetting when using fine-tuning over AL query rounds. We then develop a new class of techniques, circumventing this problem, by biasing further training towards previously labeled sets. We accomplish this by employing existing, and developing novel, replay-based Continual Learning (CL) algorithms that are effective"},"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":"2305.06408","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2023-05-10T18:34:05Z","cross_cats_sorted":[],"title_canon_sha256":"47d4f89fde4b2ff095604fbcb242f5e998f600a40686a35228ede819e0a2799e","abstract_canon_sha256":"4fc86aadd1c04ebc7918b38ca31b288797b4fa9c42bdad4dbf6132b671deb275"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:23:26.111471Z","signature_b64":"6OV4l9JyghKIs00AW+HIxeX9tJReCTZE4NEuBeSN7ul2WW0ii1ntDXZMYX2xPf+B8lZ/GQfQzkglS47MAQMECw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"4792d766f283c7c1087b6bb6f1fc479bf9652c1e48a56d3a7a27938212f4eeea","last_reissued_at":"2026-07-05T07:23:26.110924Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:23:26.110924Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Accelerating Batch Active Learning Using Continual Learning Techniques","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Arnav Das, Gantavya Bhatt, Jeff Bilmes, Megh Bhalerao, Rui Yang, Vianne Gao","submitted_at":"2023-05-10T18:34:05Z","abstract_excerpt":"A major problem with Active Learning (AL) is high training costs since models are typically retrained from scratch after every query round. We start by demonstrating that standard AL on neural networks with warm starting fails, both to accelerate training and to avoid catastrophic forgetting when using fine-tuning over AL query rounds. We then develop a new class of techniques, circumventing this problem, by biasing further training towards previously labeled sets. We accomplish this by employing existing, and developing novel, replay-based Continual Learning (CL) algorithms that are effective"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2305.06408","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/2305.06408/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":"2305.06408","created_at":"2026-07-05T07:23:26.110986+00:00"},{"alias_kind":"arxiv_version","alias_value":"2305.06408v2","created_at":"2026-07-05T07:23:26.110986+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2305.06408","created_at":"2026-07-05T07:23:26.110986+00:00"},{"alias_kind":"pith_short_12","alias_value":"I6JNOZXSQPD4","created_at":"2026-07-05T07:23:26.110986+00:00"},{"alias_kind":"pith_short_16","alias_value":"I6JNOZXSQPD4CCD3","created_at":"2026-07-05T07:23:26.110986+00:00"},{"alias_kind":"pith_short_8","alias_value":"I6JNOZXS","created_at":"2026-07-05T07:23:26.110986+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/I6JNOZXSQPD4CCD3NO3PD7CHTP","json":"https://pith.science/pith/I6JNOZXSQPD4CCD3NO3PD7CHTP.json","graph_json":"https://pith.science/api/pith-number/I6JNOZXSQPD4CCD3NO3PD7CHTP/graph.json","events_json":"https://pith.science/api/pith-number/I6JNOZXSQPD4CCD3NO3PD7CHTP/events.json","paper":"https://pith.science/paper/I6JNOZXS"},"agent_actions":{"view_html":"https://pith.science/pith/I6JNOZXSQPD4CCD3NO3PD7CHTP","download_json":"https://pith.science/pith/I6JNOZXSQPD4CCD3NO3PD7CHTP.json","view_paper":"https://pith.science/paper/I6JNOZXS","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2305.06408&json=true","fetch_graph":"https://pith.science/api/pith-number/I6JNOZXSQPD4CCD3NO3PD7CHTP/graph.json","fetch_events":"https://pith.science/api/pith-number/I6JNOZXSQPD4CCD3NO3PD7CHTP/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/I6JNOZXSQPD4CCD3NO3PD7CHTP/action/timestamp_anchor","attest_storage":"https://pith.science/pith/I6JNOZXSQPD4CCD3NO3PD7CHTP/action/storage_attestation","attest_author":"https://pith.science/pith/I6JNOZXSQPD4CCD3NO3PD7CHTP/action/author_attestation","sign_citation":"https://pith.science/pith/I6JNOZXSQPD4CCD3NO3PD7CHTP/action/citation_signature","submit_replication":"https://pith.science/pith/I6JNOZXSQPD4CCD3NO3PD7CHTP/action/replication_record"}},"created_at":"2026-07-05T07:23:26.110986+00:00","updated_at":"2026-07-05T07:23:26.110986+00:00"}