{"id":"https://openalex.org/W4401857048","doi":"https://doi.org/10.1145/3637528.3671985","title":"Approximating Memorization Using Loss Surface Geometry for Dataset Pruning and Summarization","display_name":"Approximating Memorization Using Loss Surface Geometry for Dataset Pruning and Summarization","publication_year":2024,"publication_date":"2024-08-24","ids":{"openalex":"https://openalex.org/W4401857048","doi":"https://doi.org/10.1145/3637528.3671985"},"language":"en","primary_location":{"id":"doi:10.1145/3637528.3671985","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3637528.3671985","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3637528.3671985","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 30th ACM SIGKDD Conference on Knowledge Discovery and Data Mining","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://dl.acm.org/doi/pdf/10.1145/3637528.3671985","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5053171685","display_name":"Andrea Agiollo","orcid":"https://orcid.org/0000-0003-0531-1978"},"institutions":[{"id":"https://openalex.org/I9360294","display_name":"University of Bologna","ror":"https://ror.org/01111rn36","country_code":"IT","type":"education","lineage":["https://openalex.org/I9360294"]}],"countries":["IT"],"is_corresponding":true,"raw_author_name":"Andrea Agiollo","raw_affiliation_strings":["University of Bologna, Bologna, Italy"],"affiliations":[{"raw_affiliation_string":"University of Bologna, Bologna, Italy","institution_ids":["https://openalex.org/I9360294"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5023390886","display_name":"Young In Kim","orcid":"https://orcid.org/0009-0000-5690-5675"},"institutions":[{"id":"https://openalex.org/I219193219","display_name":"Purdue University West Lafayette","ror":"https://ror.org/02dqehb95","country_code":"US","type":"education","lineage":["https://openalex.org/I219193219"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Young In Kim","raw_affiliation_strings":["Purdue University, West Lafayette, IN, USA"],"affiliations":[{"raw_affiliation_string":"Purdue University, West Lafayette, IN, USA","institution_ids":["https://openalex.org/I219193219"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5045284674","display_name":"Rajiv Khanna","orcid":"https://orcid.org/0000-0003-1314-3126"},"institutions":[{"id":"https://openalex.org/I219193219","display_name":"Purdue University West Lafayette","ror":"https://ror.org/02dqehb95","country_code":"US","type":"education","lineage":["https://openalex.org/I219193219"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Rajiv Khanna","raw_affiliation_strings":["Purdue University, West Lafayette, IN, USA"],"affiliations":[{"raw_affiliation_string":"Purdue University, West Lafayette, IN, USA","institution_ids":["https://openalex.org/I219193219"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":3,"corresponding_author_ids":["https://openalex.org/A5053171685"],"corresponding_institution_ids":["https://openalex.org/I9360294"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.13084662,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"17","last_page":"28"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10775","display_name":"Generative Adversarial Networks and Image Synthesis","score":0.9988999962806702,"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"}},"topics":[{"id":"https://openalex.org/T10775","display_name":"Generative Adversarial Networks and Image Synthesis","score":0.9988999962806702,"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"}},{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.9987999796867371,"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"}},{"id":"https://openalex.org/T12535","display_name":"Machine Learning and Data Classification","score":0.9973000288009644,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/automatic-summarization","display_name":"Automatic summarization","score":0.8342742919921875},{"id":"https://openalex.org/keywords/memorization","display_name":"Memorization","score":0.787825345993042},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.7464399933815002},{"id":"https://openalex.org/keywords/machine-learning","display_name":"Machine learning","score":0.5732730031013489},{"id":"https://openalex.org/keywords/stochastic-gradient-descent","display_name":"Stochastic gradient descent","score":0.559974193572998},{"id":"https://openalex.org/keywords/metric","display_name":"Metric (unit)","score":0.5262452960014343},{"id":"https://openalex.org/keywords/pruning","display_name":"Pruning","score":0.5081524848937988},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.4937942922115326},{"id":"https://openalex.org/keywords/minification","display_name":"Minification","score":0.45578211545944214},{"id":"https://openalex.org/keywords/matching","display_name":"Matching (statistics)","score":0.43808266520500183},{"id":"https://openalex.org/keywords/gradient-descent","display_name":"Gradient descent","score":0.43379080295562744},{"id":"https://openalex.org/keywords/data-mining","display_name":"Data mining","score":0.3940729796886444},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.3502536118030548},{"id":"https://openalex.org/keywords/pattern-recognition","display_name":"Pattern recognition (psychology)","score":0.33795422315597534},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.1576213538646698},{"id":"https://openalex.org/keywords/statistics","display_name":"Statistics","score":0.15195849537849426}],"concepts":[{"id":"https://openalex.org/C170858558","wikidata":"https://www.wikidata.org/wiki/Q1394144","display_name":"Automatic summarization","level":2,"score":0.8342742919921875},{"id":"https://openalex.org/C30038468","wikidata":"https://www.wikidata.org/wiki/Q4354775","display_name":"Memorization","level":2,"score":0.787825345993042},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7464399933815002},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.5732730031013489},{"id":"https://openalex.org/C206688291","wikidata":"https://www.wikidata.org/wiki/Q7617819","display_name":"Stochastic gradient descent","level":3,"score":0.559974193572998},{"id":"https://openalex.org/C176217482","wikidata":"https://www.wikidata.org/wiki/Q860554","display_name":"Metric (unit)","level":2,"score":0.5262452960014343},{"id":"https://openalex.org/C108010975","wikidata":"https://www.wikidata.org/wiki/Q500094","display_name":"Pruning","level":2,"score":0.5081524848937988},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4937942922115326},{"id":"https://openalex.org/C147764199","wikidata":"https://www.wikidata.org/wiki/Q6865248","display_name":"Minification","level":2,"score":0.45578211545944214},{"id":"https://openalex.org/C165064840","wikidata":"https://www.wikidata.org/wiki/Q1321061","display_name":"Matching (statistics)","level":2,"score":0.43808266520500183},{"id":"https://openalex.org/C153258448","wikidata":"https://www.wikidata.org/wiki/Q1199743","display_name":"Gradient descent","level":3,"score":0.43379080295562744},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.3940729796886444},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.3502536118030548},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.33795422315597534},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.1576213538646698},{"id":"https://openalex.org/C105795698","wikidata":"https://www.wikidata.org/wiki/Q12483","display_name":"Statistics","level":1,"score":0.15195849537849426},{"id":"https://openalex.org/C86803240","wikidata":"https://www.wikidata.org/wiki/Q420","display_name":"Biology","level":0,"score":0.0},{"id":"https://openalex.org/C162324750","wikidata":"https://www.wikidata.org/wiki/Q8134","display_name":"Economics","level":0,"score":0.0},{"id":"https://openalex.org/C199360897","wikidata":"https://www.wikidata.org/wiki/Q9143","display_name":"Programming language","level":1,"score":0.0},{"id":"https://openalex.org/C145420912","wikidata":"https://www.wikidata.org/wiki/Q853077","display_name":"Mathematics education","level":1,"score":0.0},{"id":"https://openalex.org/C21547014","wikidata":"https://www.wikidata.org/wiki/Q1423657","display_name":"Operations management","level":1,"score":0.0},{"id":"https://openalex.org/C6557445","wikidata":"https://www.wikidata.org/wiki/Q173113","display_name":"Agronomy","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3637528.3671985","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3637528.3671985","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3637528.3671985","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 30th ACM SIGKDD Conference on Knowledge Discovery and Data Mining","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.1145/3637528.3671985","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3637528.3671985","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3637528.3671985","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 30th ACM SIGKDD Conference on Knowledge Discovery and Data Mining","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4401857048.pdf"},"referenced_works_count":19,"referenced_works":["https://openalex.org/W74727102","https://openalex.org/W398859631","https://openalex.org/W1858542512","https://openalex.org/W2693098724","https://openalex.org/W2747329762","https://openalex.org/W2760000637","https://openalex.org/W3022256936","https://openalex.org/W3035261884","https://openalex.org/W3086583298","https://openalex.org/W3163841364","https://openalex.org/W3205626500","https://openalex.org/W4206648492","https://openalex.org/W4212774754","https://openalex.org/W4281693361","https://openalex.org/W4298189948","https://openalex.org/W4385412194","https://openalex.org/W4386065727","https://openalex.org/W4386072234","https://openalex.org/W4387415233"],"related_works":["https://openalex.org/W4206903459","https://openalex.org/W2754816816","https://openalex.org/W4366280654","https://openalex.org/W3160167280","https://openalex.org/W4231621013","https://openalex.org/W4362706668","https://openalex.org/W3008318776","https://openalex.org/W2041416246","https://openalex.org/W3020853991","https://openalex.org/W3035836947"],"abstract_inverted_index":{"The":[0],"sustainable":[1],"training":[2,29,48],"of":[3,22,40,44,51,159],"modern":[4],"neural":[5],"network":[6],"models":[7],"represents":[8],"an":[9],"open":[10],"challenge.":[11],"Several":[12],"existing":[13],"methods":[14],"approach":[15],"this":[16,87],"issue":[17],"by":[18],"identifying":[19],"a":[20,91,146],"subset":[21,53,76],"relevant":[23],"data":[24,30,47,54,107,143,178,188],"samples":[25],"from":[26],"the":[27,38,42,45,52,75,96,134,157],"full":[28,46],"to":[31,62,105,151],"be":[32,191],"used":[33],"in":[34],"model":[35],"optimization":[36],"with":[37,49],"goal":[39],"matching":[41],"performance":[43,181],"that":[50,70,94],"training.":[55],"Our":[56],"work":[57],"explores":[58],"using":[59],"memorization":[60,81,120],"scores":[61,82,121],"find":[63],"representative":[64],"and":[65,101,130,165,183],"atypical":[66,152,177,187],"samples.":[67],"We":[68,110,132],"demonstrate":[69],"memorization-aware":[71],"dataset":[72,163],"summarization":[73,166],"improves":[74],"construction":[77],"performance.":[78],"However,":[79],"computing":[80],"is":[83],"notably":[84],"resource-intensive.":[85],"To":[86],"end,":[88],"we":[89],"propose":[90],"novel":[92],"method":[93],"leverages":[95],"discrepancy":[97],"between":[98],"sharpness-aware":[99],"minimization":[100],"stochastic":[102],"gradient":[103],"descent":[104],"capture":[106],"points":[108,144,179,189],"atypicality.":[109],"evaluate":[111],"our":[112,137,160],"metric":[113],"over":[114],"several":[115],"efficient":[116],"approximation":[117,138],"functions":[118],"for":[119,162],"-":[122,175,182],"namely":[123],"proxies":[124],"-,":[125],"empirically":[126],"showing":[127],"superior":[128],"correlation":[129],"effectiveness.":[131],"explore":[133],"causes":[135],"behind":[136],"quality,":[139],"highlighting":[140],"how":[141],"typical":[142],"trigger":[145],"flatter":[147],"loss":[148],"landscape":[149],"compared":[150],"ones.":[153],"Extensive":[154],"experiments":[155],"confirm":[156],"effectiveness":[158],"proxy":[161],"pruning":[164],"tasks,":[167],"surpassing":[168],"state-of-the-art":[169],"approaches":[170],"both":[171],"on":[172],"canonical":[173],"setups":[174],"where":[176],"benefit":[180],"few-shot":[184],"learning":[185],"scenarios-where":[186],"can":[190],"detrimental.":[192]},"counts_by_year":[],"updated_date":"2025-11-06T03:46:38.306776","created_date":"2025-10-10T00:00:00"}