{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:P4S33OVCERXFTHCOA3NBLGF3MA","short_pith_number":"pith:P4S33OVC","schema_version":"1.0","canonical_sha256":"7f25bdbaa2246e599c4e06da1598bb601ef7e69ac81c9166c73be49f7d3d6b4d","source":{"kind":"arxiv","id":"2310.06511","version":3},"attestation_state":"computed","paper":{"title":"Self-Supervised Dataset Distillation for Transfer Learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Dong Bok Lee, Joonho Ko, Juho Lee, Kenji Kawaguchi, Seanie Lee, Sung Ju Hwang","submitted_at":"2023-10-10T10:48:52Z","abstract_excerpt":"Dataset distillation methods have achieved remarkable success in distilling a large dataset into a small set of representative samples. However, they are not designed to produce a distilled dataset that can be effectively used for facilitating self-supervised pre-training. To this end, we propose a novel problem of distilling an unlabeled dataset into a set of small synthetic samples for efficient self-supervised learning (SSL). We first prove that a gradient of synthetic samples with respect to a SSL objective in naive bilevel optimization is \\textit{biased} due to the randomness originating "},"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":"2310.06511","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2023-10-10T10:48:52Z","cross_cats_sorted":[],"title_canon_sha256":"8fa248f6b7624ee3225eb54cf6d56d223ab7dcb85ac3bee3d83fa4baf08c9f9d","abstract_canon_sha256":"295d56478da4d791c0b327992888b9c50efc556bf5c95d3bd3ea3ccc65a08836"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:07:11.527821Z","signature_b64":"L6g1m1PzE+Ch1+f3EbCZCIQk6k8ocaKDW4rC8HXE3MNY5f2h56iDYBsmgSBwiGB4kOl/kYKhcRglPaQ23mHODA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7f25bdbaa2246e599c4e06da1598bb601ef7e69ac81c9166c73be49f7d3d6b4d","last_reissued_at":"2026-07-05T08:07:11.527350Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:07:11.527350Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Self-Supervised Dataset Distillation for Transfer Learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Dong Bok Lee, Joonho Ko, Juho Lee, Kenji Kawaguchi, Seanie Lee, Sung Ju Hwang","submitted_at":"2023-10-10T10:48:52Z","abstract_excerpt":"Dataset distillation methods have achieved remarkable success in distilling a large dataset into a small set of representative samples. However, they are not designed to produce a distilled dataset that can be effectively used for facilitating self-supervised pre-training. To this end, we propose a novel problem of distilling an unlabeled dataset into a set of small synthetic samples for efficient self-supervised learning (SSL). We first prove that a gradient of synthetic samples with respect to a SSL objective in naive bilevel optimization is \\textit{biased} due to the randomness originating "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2310.06511","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/2310.06511/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":"2310.06511","created_at":"2026-07-05T08:07:11.527408+00:00"},{"alias_kind":"arxiv_version","alias_value":"2310.06511v3","created_at":"2026-07-05T08:07:11.527408+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2310.06511","created_at":"2026-07-05T08:07:11.527408+00:00"},{"alias_kind":"pith_short_12","alias_value":"P4S33OVCERXF","created_at":"2026-07-05T08:07:11.527408+00:00"},{"alias_kind":"pith_short_16","alias_value":"P4S33OVCERXFTHCO","created_at":"2026-07-05T08:07:11.527408+00:00"},{"alias_kind":"pith_short_8","alias_value":"P4S33OVC","created_at":"2026-07-05T08:07:11.527408+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/P4S33OVCERXFTHCOA3NBLGF3MA","json":"https://pith.science/pith/P4S33OVCERXFTHCOA3NBLGF3MA.json","graph_json":"https://pith.science/api/pith-number/P4S33OVCERXFTHCOA3NBLGF3MA/graph.json","events_json":"https://pith.science/api/pith-number/P4S33OVCERXFTHCOA3NBLGF3MA/events.json","paper":"https://pith.science/paper/P4S33OVC"},"agent_actions":{"view_html":"https://pith.science/pith/P4S33OVCERXFTHCOA3NBLGF3MA","download_json":"https://pith.science/pith/P4S33OVCERXFTHCOA3NBLGF3MA.json","view_paper":"https://pith.science/paper/P4S33OVC","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2310.06511&json=true","fetch_graph":"https://pith.science/api/pith-number/P4S33OVCERXFTHCOA3NBLGF3MA/graph.json","fetch_events":"https://pith.science/api/pith-number/P4S33OVCERXFTHCOA3NBLGF3MA/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/P4S33OVCERXFTHCOA3NBLGF3MA/action/timestamp_anchor","attest_storage":"https://pith.science/pith/P4S33OVCERXFTHCOA3NBLGF3MA/action/storage_attestation","attest_author":"https://pith.science/pith/P4S33OVCERXFTHCOA3NBLGF3MA/action/author_attestation","sign_citation":"https://pith.science/pith/P4S33OVCERXFTHCOA3NBLGF3MA/action/citation_signature","submit_replication":"https://pith.science/pith/P4S33OVCERXFTHCOA3NBLGF3MA/action/replication_record"}},"created_at":"2026-07-05T08:07:11.527408+00:00","updated_at":"2026-07-05T08:07:11.527408+00:00"}