{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:KIGEMZXJ6WPCCLMAU2MRHOUUVT","short_pith_number":"pith:KIGEMZXJ","schema_version":"1.0","canonical_sha256":"520c4666e9f59e212d80a69913ba94acff4d5bf3b16e73a4e8882f7ccddcddb2","source":{"kind":"arxiv","id":"2502.11756","version":1},"attestation_state":"computed","paper":{"title":"On the Computation of the Fisher Information in Continual Learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CV","stat.ML"],"primary_cat":"cs.LG","authors_text":"Gido M. van de Ven","submitted_at":"2025-02-17T12:52:10Z","abstract_excerpt":"One of the most popular methods for continual learning with deep neural networks is Elastic Weight Consolidation (EWC), which involves computing the Fisher Information. The exact way in which the Fisher Information is computed is however rarely described, and multiple different implementations for it can be found online. This blog post discusses and empirically compares several often-used implementations, which highlights that many currently reported results for EWC could likely be improved by changing the way the Fisher Information is computed."},"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":"2502.11756","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-02-17T12:52:10Z","cross_cats_sorted":["cs.AI","cs.CV","stat.ML"],"title_canon_sha256":"9f50b4e0d4dfbf49e1665e69ca4bbe5f9a57541e1958a531a08e4f2936d9e891","abstract_canon_sha256":"cbe551e9e6f70fd83d8abf103147de3a531a4949e18d3754bc39cc24e6d67d11"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:15:33.686555Z","signature_b64":"y51LcqH4R2aLAjTNWJk4/Y1Et8VebrG5vAWU0j+CJ2bQPSS0LVLaUnOW9IfjcmzRw1X7hmuersB6ChB04XNHBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"520c4666e9f59e212d80a69913ba94acff4d5bf3b16e73a4e8882f7ccddcddb2","last_reissued_at":"2026-07-05T10:15:33.686102Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:15:33.686102Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"On the Computation of the Fisher Information in Continual Learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CV","stat.ML"],"primary_cat":"cs.LG","authors_text":"Gido M. van de Ven","submitted_at":"2025-02-17T12:52:10Z","abstract_excerpt":"One of the most popular methods for continual learning with deep neural networks is Elastic Weight Consolidation (EWC), which involves computing the Fisher Information. The exact way in which the Fisher Information is computed is however rarely described, and multiple different implementations for it can be found online. This blog post discusses and empirically compares several often-used implementations, which highlights that many currently reported results for EWC could likely be improved by changing the way the Fisher Information is computed."},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.11756","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/2502.11756/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":"2502.11756","created_at":"2026-07-05T10:15:33.686169+00:00"},{"alias_kind":"arxiv_version","alias_value":"2502.11756v1","created_at":"2026-07-05T10:15:33.686169+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.11756","created_at":"2026-07-05T10:15:33.686169+00:00"},{"alias_kind":"pith_short_12","alias_value":"KIGEMZXJ6WPC","created_at":"2026-07-05T10:15:33.686169+00:00"},{"alias_kind":"pith_short_16","alias_value":"KIGEMZXJ6WPCCLMA","created_at":"2026-07-05T10:15:33.686169+00:00"},{"alias_kind":"pith_short_8","alias_value":"KIGEMZXJ","created_at":"2026-07-05T10:15:33.686169+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":4,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2508.18187","citing_title":"BRAIN: Bias-Mitigation Continual Learning Approach to Vision-Brain Understanding","ref_index":105,"is_internal_anchor":false},{"citing_arxiv_id":"2603.20410","citing_title":"SLE-FNO: Single-Layer Extensions for Task-Agnostic Continual Learning in Fourier Neural Operators","ref_index":100,"is_internal_anchor":false},{"citing_arxiv_id":"2605.07076","citing_title":"Self-Consolidating Language Models: Continual Knowledge Incorporation from Context","ref_index":16,"is_internal_anchor":false},{"citing_arxiv_id":"2605.07076","citing_title":"Self-Consolidating Language Models: Continual Knowledge Incorporation from Context","ref_index":16,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/KIGEMZXJ6WPCCLMAU2MRHOUUVT","json":"https://pith.science/pith/KIGEMZXJ6WPCCLMAU2MRHOUUVT.json","graph_json":"https://pith.science/api/pith-number/KIGEMZXJ6WPCCLMAU2MRHOUUVT/graph.json","events_json":"https://pith.science/api/pith-number/KIGEMZXJ6WPCCLMAU2MRHOUUVT/events.json","paper":"https://pith.science/paper/KIGEMZXJ"},"agent_actions":{"view_html":"https://pith.science/pith/KIGEMZXJ6WPCCLMAU2MRHOUUVT","download_json":"https://pith.science/pith/KIGEMZXJ6WPCCLMAU2MRHOUUVT.json","view_paper":"https://pith.science/paper/KIGEMZXJ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2502.11756&json=true","fetch_graph":"https://pith.science/api/pith-number/KIGEMZXJ6WPCCLMAU2MRHOUUVT/graph.json","fetch_events":"https://pith.science/api/pith-number/KIGEMZXJ6WPCCLMAU2MRHOUUVT/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/KIGEMZXJ6WPCCLMAU2MRHOUUVT/action/timestamp_anchor","attest_storage":"https://pith.science/pith/KIGEMZXJ6WPCCLMAU2MRHOUUVT/action/storage_attestation","attest_author":"https://pith.science/pith/KIGEMZXJ6WPCCLMAU2MRHOUUVT/action/author_attestation","sign_citation":"https://pith.science/pith/KIGEMZXJ6WPCCLMAU2MRHOUUVT/action/citation_signature","submit_replication":"https://pith.science/pith/KIGEMZXJ6WPCCLMAU2MRHOUUVT/action/replication_record"}},"created_at":"2026-07-05T10:15:33.686169+00:00","updated_at":"2026-07-05T10:15:33.686169+00:00"}