{"id":"https://openalex.org/W7220994239","doi":"https://doi.org/10.48550/arxiv.2610.10347","title":"Dataset Pruning from First Principles: A Label-Free Linear Programming Approach","display_name":"Dataset Pruning from First Principles: A Label-Free Linear Programming Approach","publication_year":2026,"publication_date":"2026-10-07","ids":{"openalex":"https://openalex.org/W7220994239","doi":"https://doi.org/10.48550/arxiv.2610.10347"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2610.10347","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2610.10347","pdf_url":"https://arxiv.org/pdf/2610.10347","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":"other-oa","license_id":"https://openalex.org/licenses/other-oa","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"volume":null,"type":"preprint","indexed_in":["arxiv","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2610.10347","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5137325243","display_name":"Rodrigo Schuller","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I141883831","display_name":"Instituto Nacional de Matem\u00e1tica Pura e Aplicada","ror":"https://ror.org/028caqe42","country_code":"BR","type":"facility","lineage":["https://openalex.org/I141883831","https://openalex.org/I4210151455"]}],"countries":["BR"],"is_corresponding":false,"raw_author_name":"Schuller, Rodrigo","raw_affiliation_strings":["Instituto de Matem\u00e1tica Pura e Aplicada (IMPA) Rio de Janeiro , Brazil"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Instituto de Matem\u00e1tica Pura e Aplicada (IMPA) Rio de Janeiro , Brazil","institution_ids":["https://openalex.org/I141883831"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5157976787","display_name":"Francisco Ganacim","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I141883831","display_name":"Instituto Nacional de Matem\u00e1tica Pura e Aplicada","ror":"https://ror.org/028caqe42","country_code":"BR","type":"facility","lineage":["https://openalex.org/I141883831","https://openalex.org/I4210151455"]}],"countries":["BR"],"is_corresponding":false,"raw_author_name":"Ganacim, Francisco","raw_affiliation_strings":["Instituto de Matem\u00e1tica Pura e Aplicada (IMPA) Rio de Janeiro , Brazil"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Instituto de Matem\u00e1tica Pura e Aplicada (IMPA) Rio de Janeiro , Brazil","institution_ids":["https://openalex.org/I141883831"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I141883831"],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"cited_by_count":0,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":null,"last_page":null},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T12535","display_name":"Machine Learning and Data Classification","score":0.8666999936103821,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},"topics":[{"id":"https://openalex.org/T12535","display_name":"Machine Learning and Data Classification","score":0.8666999936103821,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T11612","display_name":"Stochastic Gradient Optimization Techniques","score":0.12870000302791595,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.0010999999940395355,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/dataset-pruning","display_name":"dataset pruning","score":0.9900000095367432},{"id":"https://openalex.org/keywords/variance-minimization","display_name":"variance minimization","score":0.8659999966621399},{"id":"https://openalex.org/keywords/linear-programming","display_name":"linear programming","score":0.8429999947547913},{"id":"https://openalex.org/keywords/rigid-body-motion","display_name":"rigid body motion","score":0.5860000252723694},{"id":"https://openalex.org/keywords/training-set-reduction","display_name":"training data reduction","score":0.5210000276565552}],"concepts":[{"id":"https://openalex.org/C108010975","wikidata":"https://www.wikidata.org/wiki/Q500094","display_name":"Pruning","level":2,"score":0.6658999919891357},{"id":"https://openalex.org/C184898388","wikidata":"https://www.wikidata.org/wiki/Q1435712","display_name":"Pairwise comparison","level":2,"score":0.661899983882904},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.6312999725341797},{"id":"https://openalex.org/C81917197","wikidata":"https://www.wikidata.org/wiki/Q628760","display_name":"Selection (genetic algorithm)","level":2,"score":0.5809999704360962},{"id":"https://openalex.org/C140779682","wikidata":"https://www.wikidata.org/wiki/Q210868","display_name":"Sampling (signal processing)","level":3,"score":0.5285999774932861},{"id":"https://openalex.org/C41045048","wikidata":"https://www.wikidata.org/wiki/Q202843","display_name":"Linear programming","level":2,"score":0.5109000205993652},{"id":"https://openalex.org/C196083921","wikidata":"https://www.wikidata.org/wiki/Q7915758","display_name":"Variance (accounting)","level":2,"score":0.47380000352859497},{"id":"https://openalex.org/C126255220","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Mathematical optimization","level":1,"score":0.4706000089645386},{"id":"https://openalex.org/C147764199","wikidata":"https://www.wikidata.org/wiki/Q6865248","display_name":"Minification","level":2,"score":0.4323999881744385},{"id":"https://openalex.org/C177264268","wikidata":"https://www.wikidata.org/wiki/Q1514741","display_name":"Set (abstract data type)","level":2,"score":0.4185999929904938},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.41519999504089355},{"id":"https://openalex.org/C41608201","wikidata":"https://www.wikidata.org/wiki/Q980509","display_name":"Embedding","level":2,"score":0.4034999907016754},{"id":"https://openalex.org/C103545067","wikidata":"https://www.wikidata.org/wiki/Q796265","display_name":"Best linear unbiased prediction","level":3,"score":0.3774999976158142},{"id":"https://openalex.org/C163175372","wikidata":"https://www.wikidata.org/wiki/Q3339222","display_name":"Linear model","level":2,"score":0.3452000021934509},{"id":"https://openalex.org/C93959086","wikidata":"https://www.wikidata.org/wiki/Q6888345","display_name":"Model selection","level":2,"score":0.3416000008583069},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.3244999945163727},{"id":"https://openalex.org/C52740198","wikidata":"https://www.wikidata.org/wiki/Q1539564","display_name":"Importance sampling","level":3,"score":0.31619998812675476},{"id":"https://openalex.org/C80899671","wikidata":"https://www.wikidata.org/wiki/Q1304193","display_name":"Vertex (graph theory)","level":3,"score":0.29789999127388},{"id":"https://openalex.org/C311688","wikidata":"https://www.wikidata.org/wiki/Q2393193","display_name":"Time complexity","level":2,"score":0.28850001096725464},{"id":"https://openalex.org/C105795698","wikidata":"https://www.wikidata.org/wiki/Q12483","display_name":"Statistics","level":1,"score":0.27970001101493835},{"id":"https://openalex.org/C82152865","wikidata":"https://www.wikidata.org/wiki/Q7208505","display_name":"Poisson sampling","level":5,"score":0.26980000734329224},{"id":"https://openalex.org/C183905921","wikidata":"https://www.wikidata.org/wiki/Q1038757","display_name":"Multiple comparisons problem","level":2,"score":0.26750001311302185},{"id":"https://openalex.org/C165646398","wikidata":"https://www.wikidata.org/wiki/Q3755281","display_name":"Minimum-variance unbiased estimator","level":3,"score":0.26499998569488525},{"id":"https://openalex.org/C148764684","wikidata":"https://www.wikidata.org/wiki/Q621751","display_name":"Approximation algorithm","level":2,"score":0.2644999921321869},{"id":"https://openalex.org/C2777303404","wikidata":"https://www.wikidata.org/wiki/Q759757","display_name":"Convergence (economics)","level":2,"score":0.2623000144958496},{"id":"https://openalex.org/C21080849","wikidata":"https://www.wikidata.org/wiki/Q13611879","display_name":"Data point","level":2,"score":0.258899986743927},{"id":"https://openalex.org/C81845259","wikidata":"https://www.wikidata.org/wiki/Q290117","display_name":"Quadratic programming","level":2,"score":0.257999986410141},{"id":"https://openalex.org/C58489278","wikidata":"https://www.wikidata.org/wiki/Q1172284","display_name":"Data set","level":2,"score":0.2538999915122986},{"id":"https://openalex.org/C145691206","wikidata":"https://www.wikidata.org/wiki/Q747980","display_name":"Polytope","level":2,"score":0.2526000142097473}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2610.10347","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2610.10347","pdf_url":"https://arxiv.org/pdf/2610.10347","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":"other-oa","license_id":"https://openalex.org/licenses/other-oa","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"doi:10.48550/arxiv.2610.10347","doi":"https://doi.org/10.48550/arxiv.2610.10347","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2610.10347","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2610.10347","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2610.10347","pdf_url":"https://arxiv.org/pdf/2610.10347","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":"other-oa","license_id":"https://openalex.org/licenses/other-oa","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"sustainable_development_goals":[],"study_designs":[],"awards":[],"funders":[{"id":"https://openalex.org/F4320322025","display_name":"Conselho Nacional de Desenvolvimento Cient\u00edfico e Tecnol\u00f3gico","ror":"https://ror.org/03swz6y49"}],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7220994239.pdf","grobid_xml":"https://content.openalex.org/works/W7220994239.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W4281258849","https://openalex.org/W7164806843","https://openalex.org/W7164536056","https://openalex.org/W4414286474","https://openalex.org/W7160862941","https://openalex.org/W4402915941","https://openalex.org/W4416628744","https://openalex.org/W4407423360","https://openalex.org/W4417228085","https://openalex.org/W7214612477"],"abstract_inverted_index":{"Dataset":[0],"pruning":[1],"reduces":[2,193],"a":[3,8,45,71,79,91,142],"large":[4],"training":[5,150],"set":[6],"to":[7,124],"representative":[9],"subset":[10,42,53,75],"while":[11,138,201],"preserving":[12,139],"model":[13,66,149],"performance.":[14],"Existing":[15],"geometry-based":[16],"methods":[17,178],"typically":[18],"assume":[19],"that":[20,51,130,144],"nearby":[21],"points":[22],"in":[23,58,95,166,179],"embedding":[24],"space":[25],"share":[26],"similar":[27],"properties.":[28],"Rather":[29],"than":[30],"imposing":[31],"this":[32,83],"assumption,":[33],"we":[34,69],"derive":[35],"geometric":[36,177],"selection":[37,43,76,185],"criteria":[38],"by":[39,196],"reformulating":[40],"unbiased":[41,74],"as":[44,78],"variance":[46,89,136,195],"minimization":[47],"problem.":[48],"Unbiasedness":[49],"ensures":[50],"unweighted":[52],"averages":[54,57],"recover":[55],"full-dataset":[56],"expectation,":[59],"including":[60],"losses":[61],"and":[62,156,174],"gradients":[63],"at":[64,170,183],"fixed":[65],"parameters.":[67],"Specifically,":[68],"characterize":[70],"family":[72],"of":[73,134],"algorithms":[77],"high-dimensional":[80],"polytope.":[81],"In":[82],"context,":[84],"minimizing":[85],"the":[86,107,135,190,203],"expected":[87],"sampling":[88,96,165],"is":[90,117],"linear":[92,115],"objective.":[93],"Differences":[94],"variance,":[97],"averaged":[98],"over":[99],"rigid":[100],"motions,":[101],"admit":[102],"closed-form":[103],"pairwise":[104],"expressions.":[105],"Because":[106],"polytope":[108],"has":[109],"high":[110],"dimension,":[111],"directly":[112],"applying":[113],"standard":[114],"programming":[116],"impractical.":[118],"We":[119],"instead":[120],"use":[121],"these":[122],"expressions":[123],"construct":[125],"an":[126,132],"efficient":[127],"vertex":[128],"walk":[129],"optimizes":[131],"approximation":[133],"objective":[137],"unbiasedness,":[140],"yielding":[141],"method":[143,160],"requires":[145],"neither":[146],"labels":[147],"nor":[148],"during":[151],"selection.":[152],"Across":[153],"CIFAR-10,":[154],"MNIST,":[155],"CelebA":[157],"benchmarks,":[158],"our":[159],"matches":[161],"or":[162],"exceeds":[163],"uniform":[164],"mean":[167],"test":[168],"accuracy":[169],"every":[171],"evaluated":[172],"budget":[173],"outperforms":[175],"competing":[176],"several":[180],"settings,":[181],"particularly":[182],"small":[184],"budgets.":[186],"Beyond":[187],"dataset":[188],"pruning,":[189],"same":[191],"framework":[192],"stochastic-gradient":[194],"increasing":[197],"diversity":[198],"within":[199],"mini-batches":[200],"keeping":[202],"batch":[204],"size":[205],"unchanged.":[206]},"counts_by_year":[],"updated_date":"2026-10-28T10:33:24.094632","created_date":"2026-10-09T00:00:00"}