{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:G2ILNJ2266QAW6XSRLXWOLQPBA","short_pith_number":"pith:G2ILNJ22","schema_version":"1.0","canonical_sha256":"3690b6a75af7a00b7af28aef672e0f0801e9e4565aab13d5a176f2a05b1ba00b","source":{"kind":"arxiv","id":"2505.08557","version":1},"attestation_state":"computed","paper":{"title":"Online Learning and Unlearning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Amartya Sanyal, Bernhard Sch\\\"olkopf, Yaxi Hu","submitted_at":"2025-05-13T13:33:36Z","abstract_excerpt":"We formalize the problem of online learning-unlearning, where a model is updated sequentially in an online setting while accommodating unlearning requests between updates. After a data point is unlearned, all subsequent outputs must be statistically indistinguishable from those of a model trained without that point. We present two online learner-unlearner (OLU) algorithms, both built upon online gradient descent (OGD). The first, passive OLU, leverages OGD's contractive property and injects noise when unlearning occurs, incurring no additional computation. The second, active OLU, uses an offli"},"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":"2505.08557","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-05-13T13:33:36Z","cross_cats_sorted":[],"title_canon_sha256":"bae6c834e1cf308881d788ddac24045d2ef272089030a4aa436f4a9f1c05f4c0","abstract_canon_sha256":"1d7f5ebd9c12566c26fd9e62b1bba511d85d8f78fa61e60797e8fb9527651fec"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:02:32.487758Z","signature_b64":"ATO6ITr7RGHU6CN5XxkdUKE8PsB8I8W6GdBcmszbqjrtyDGCkIJhXdrMq2M5l1jy1zAJzshcPpohzEJ1sfSlDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3690b6a75af7a00b7af28aef672e0f0801e9e4565aab13d5a176f2a05b1ba00b","last_reissued_at":"2026-07-05T11:02:32.487279Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:02:32.487279Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Online Learning and Unlearning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Amartya Sanyal, Bernhard Sch\\\"olkopf, Yaxi Hu","submitted_at":"2025-05-13T13:33:36Z","abstract_excerpt":"We formalize the problem of online learning-unlearning, where a model is updated sequentially in an online setting while accommodating unlearning requests between updates. After a data point is unlearned, all subsequent outputs must be statistically indistinguishable from those of a model trained without that point. We present two online learner-unlearner (OLU) algorithms, both built upon online gradient descent (OGD). The first, passive OLU, leverages OGD's contractive property and injects noise when unlearning occurs, incurring no additional computation. The second, active OLU, uses an offli"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.08557","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/2505.08557/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":"2505.08557","created_at":"2026-07-05T11:02:32.487342+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.08557v1","created_at":"2026-07-05T11:02:32.487342+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.08557","created_at":"2026-07-05T11:02:32.487342+00:00"},{"alias_kind":"pith_short_12","alias_value":"G2ILNJ2266QA","created_at":"2026-07-05T11:02:32.487342+00:00"},{"alias_kind":"pith_short_16","alias_value":"G2ILNJ2266QAW6XS","created_at":"2026-07-05T11:02:32.487342+00:00"},{"alias_kind":"pith_short_8","alias_value":"G2ILNJ22","created_at":"2026-07-05T11:02:32.487342+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.00638","citing_title":"Unlearning Offline Stochastic Multi-Armed Bandits","ref_index":17,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/G2ILNJ2266QAW6XSRLXWOLQPBA","json":"https://pith.science/pith/G2ILNJ2266QAW6XSRLXWOLQPBA.json","graph_json":"https://pith.science/api/pith-number/G2ILNJ2266QAW6XSRLXWOLQPBA/graph.json","events_json":"https://pith.science/api/pith-number/G2ILNJ2266QAW6XSRLXWOLQPBA/events.json","paper":"https://pith.science/paper/G2ILNJ22"},"agent_actions":{"view_html":"https://pith.science/pith/G2ILNJ2266QAW6XSRLXWOLQPBA","download_json":"https://pith.science/pith/G2ILNJ2266QAW6XSRLXWOLQPBA.json","view_paper":"https://pith.science/paper/G2ILNJ22","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.08557&json=true","fetch_graph":"https://pith.science/api/pith-number/G2ILNJ2266QAW6XSRLXWOLQPBA/graph.json","fetch_events":"https://pith.science/api/pith-number/G2ILNJ2266QAW6XSRLXWOLQPBA/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/G2ILNJ2266QAW6XSRLXWOLQPBA/action/timestamp_anchor","attest_storage":"https://pith.science/pith/G2ILNJ2266QAW6XSRLXWOLQPBA/action/storage_attestation","attest_author":"https://pith.science/pith/G2ILNJ2266QAW6XSRLXWOLQPBA/action/author_attestation","sign_citation":"https://pith.science/pith/G2ILNJ2266QAW6XSRLXWOLQPBA/action/citation_signature","submit_replication":"https://pith.science/pith/G2ILNJ2266QAW6XSRLXWOLQPBA/action/replication_record"}},"created_at":"2026-07-05T11:02:32.487342+00:00","updated_at":"2026-07-05T11:02:32.487342+00:00"}