{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2020:BMAPUZ24HME5LAF5RSNGVCT4PB","short_pith_number":"pith:BMAPUZ24","schema_version":"1.0","canonical_sha256":"0b00fa675c3b09d580bd8c9a6a8a7c78756e6ed5bde86f2d32ccf2bc506e3c75","source":{"kind":"arxiv","id":"2002.08253","version":3},"attestation_state":"computed","paper":{"title":"Distance-Based Regularisation of Deep Networks for Fine-Tuning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"stat.ML","authors_text":"Henry Gouk, Massimiliano Pontil, Timothy M. Hospedales","submitted_at":"2020-02-19T16:00:47Z","abstract_excerpt":"We investigate approaches to regularisation during fine-tuning of deep neural networks. First we provide a neural network generalisation bound based on Rademacher complexity that uses the distance the weights have moved from their initial values. This bound has no direct dependence on the number of weights and compares favourably to other bounds when applied to convolutional networks. Our bound is highly relevant for fine-tuning, because providing a network with a good initialisation based on transfer learning means that learning can modify the weights less, and hence achieve tighter generalis"},"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":"2002.08253","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"stat.ML","submitted_at":"2020-02-19T16:00:47Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"af272cf086a7fd1b3e36ab39597f30a6ccfca09c34fd42a8c71d90ba1d96bb9e","abstract_canon_sha256":"0466802c6074f92d3aaab6a75b7ed2b1691c481f52bc115dc6b854aa22eddc91"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:07:01.512215Z","signature_b64":"1jIM7fA5L6f1Pi0VazRwwwql5JkGuQ7psGRtY8riqOaP1a6JZmfFSU/hMazK/Cyw2zlfrKd4WZYkrolqGLnFAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0b00fa675c3b09d580bd8c9a6a8a7c78756e6ed5bde86f2d32ccf2bc506e3c75","last_reissued_at":"2026-07-05T02:07:01.511734Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:07:01.511734Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Distance-Based Regularisation of Deep Networks for Fine-Tuning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"stat.ML","authors_text":"Henry Gouk, Massimiliano Pontil, Timothy M. Hospedales","submitted_at":"2020-02-19T16:00:47Z","abstract_excerpt":"We investigate approaches to regularisation during fine-tuning of deep neural networks. First we provide a neural network generalisation bound based on Rademacher complexity that uses the distance the weights have moved from their initial values. This bound has no direct dependence on the number of weights and compares favourably to other bounds when applied to convolutional networks. Our bound is highly relevant for fine-tuning, because providing a network with a good initialisation based on transfer learning means that learning can modify the weights less, and hence achieve tighter generalis"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2002.08253","kind":"arxiv","version":3},"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/2002.08253/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":"2002.08253","created_at":"2026-07-05T02:07:01.511792+00:00"},{"alias_kind":"arxiv_version","alias_value":"2002.08253v3","created_at":"2026-07-05T02:07:01.511792+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2002.08253","created_at":"2026-07-05T02:07:01.511792+00:00"},{"alias_kind":"pith_short_12","alias_value":"BMAPUZ24HME5","created_at":"2026-07-05T02:07:01.511792+00:00"},{"alias_kind":"pith_short_16","alias_value":"BMAPUZ24HME5LAF5","created_at":"2026-07-05T02:07:01.511792+00:00"},{"alias_kind":"pith_short_8","alias_value":"BMAPUZ24","created_at":"2026-07-05T02:07:01.511792+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.08850","citing_title":"Local LMO: Constrained Gradient Optimization via a Local Linear Minimization Oracle","ref_index":81,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/BMAPUZ24HME5LAF5RSNGVCT4PB","json":"https://pith.science/pith/BMAPUZ24HME5LAF5RSNGVCT4PB.json","graph_json":"https://pith.science/api/pith-number/BMAPUZ24HME5LAF5RSNGVCT4PB/graph.json","events_json":"https://pith.science/api/pith-number/BMAPUZ24HME5LAF5RSNGVCT4PB/events.json","paper":"https://pith.science/paper/BMAPUZ24"},"agent_actions":{"view_html":"https://pith.science/pith/BMAPUZ24HME5LAF5RSNGVCT4PB","download_json":"https://pith.science/pith/BMAPUZ24HME5LAF5RSNGVCT4PB.json","view_paper":"https://pith.science/paper/BMAPUZ24","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2002.08253&json=true","fetch_graph":"https://pith.science/api/pith-number/BMAPUZ24HME5LAF5RSNGVCT4PB/graph.json","fetch_events":"https://pith.science/api/pith-number/BMAPUZ24HME5LAF5RSNGVCT4PB/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/BMAPUZ24HME5LAF5RSNGVCT4PB/action/timestamp_anchor","attest_storage":"https://pith.science/pith/BMAPUZ24HME5LAF5RSNGVCT4PB/action/storage_attestation","attest_author":"https://pith.science/pith/BMAPUZ24HME5LAF5RSNGVCT4PB/action/author_attestation","sign_citation":"https://pith.science/pith/BMAPUZ24HME5LAF5RSNGVCT4PB/action/citation_signature","submit_replication":"https://pith.science/pith/BMAPUZ24HME5LAF5RSNGVCT4PB/action/replication_record"}},"created_at":"2026-07-05T02:07:01.511792+00:00","updated_at":"2026-07-05T02:07:01.511792+00:00"}