{"id":"https://openalex.org/W7107869341","doi":"https://doi.org/10.1109/apsipaasc65261.2025.11249160","title":"Monomial Matrix Relocation on the Loss Function Level-Set of Feedforward Neural Networks","display_name":"Monomial Matrix Relocation on the Loss Function Level-Set of Feedforward Neural Networks","publication_year":2025,"publication_date":"2025-10-22","ids":{"openalex":"https://openalex.org/W7107869341","doi":"https://doi.org/10.1109/apsipaasc65261.2025.11249160"},"language":null,"primary_location":{"id":"doi:10.1109/apsipaasc65261.2025.11249160","is_oa":false,"landing_page_url":"https://doi.org/10.1109/apsipaasc65261.2025.11249160","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 Asia Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA ASC)","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":false,"oa_status":"closed","oa_url":null,"any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":null,"display_name":"Ozgur Soysal","orcid":null},"institutions":[{"id":"https://openalex.org/I168864056","display_name":"Bilkent University","ror":"https://ror.org/02vh8a032","country_code":"TR","type":"education","lineage":["https://openalex.org/I168864056"]}],"countries":["TR"],"is_corresponding":false,"raw_author_name":"Ozgur Soysal","raw_affiliation_strings":["Bilkent University,Department of Electrical and Electronics Engineering,Ankara,Turkey"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Bilkent University,Department of Electrical and Electronics Engineering,Ankara,Turkey","institution_ids":["https://openalex.org/I168864056"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Arda Ozdemir","orcid":null},"institutions":[{"id":"https://openalex.org/I168864056","display_name":"Bilkent University","ror":"https://ror.org/02vh8a032","country_code":"TR","type":"education","lineage":["https://openalex.org/I168864056"]}],"countries":["TR"],"is_corresponding":false,"raw_author_name":"Arda Ozdemir","raw_affiliation_strings":["Bilkent University,Department of Electrical and Electronics Engineering,Ankara,Turkey"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Bilkent University,Department of Electrical and Electronics Engineering,Ankara,Turkey","institution_ids":["https://openalex.org/I168864056"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Yigit Yildirim","orcid":null},"institutions":[{"id":"https://openalex.org/I168864056","display_name":"Bilkent University","ror":"https://ror.org/02vh8a032","country_code":"TR","type":"education","lineage":["https://openalex.org/I168864056"]}],"countries":["TR"],"is_corresponding":false,"raw_author_name":"Yigit Yildirim","raw_affiliation_strings":["Bilkent University,Department of Electrical and Electronics Engineering,Ankara,Turkey"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Bilkent University,Department of Electrical and Electronics Engineering,Ankara,Turkey","institution_ids":["https://openalex.org/I168864056"]}]},{"author_position":"last","author":{"id":null,"display_name":"Orhan Arikan","orcid":null},"institutions":[{"id":"https://openalex.org/I168864056","display_name":"Bilkent University","ror":"https://ror.org/02vh8a032","country_code":"TR","type":"education","lineage":["https://openalex.org/I168864056"]}],"countries":["TR"],"is_corresponding":false,"raw_author_name":"Orhan Arikan","raw_affiliation_strings":["Bilkent University,Department of Electrical and Electronics Engineering,Ankara,Turkey"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Bilkent University,Department of Electrical and Electronics Engineering,Ankara,Turkey","institution_ids":["https://openalex.org/I168864056"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I168864056"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.73977653,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"1188","last_page":"1193"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11612","display_name":"Stochastic Gradient Optimization Techniques","score":0.37380000948905945,"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/T11612","display_name":"Stochastic Gradient Optimization Techniques","score":0.37380000948905945,"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/T12676","display_name":"Machine Learning and ELM","score":0.07760000228881836,"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/T12535","display_name":"Machine Learning and Data Classification","score":0.050999999046325684,"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/mnist-database","display_name":"MNIST database","score":0.8521999716758728},{"id":"https://openalex.org/keywords/monomial","display_name":"Monomial","score":0.7940000295639038},{"id":"https://openalex.org/keywords/maxima-and-minima","display_name":"Maxima and minima","score":0.769599974155426},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.5831000208854675},{"id":"https://openalex.org/keywords/simulated-annealing","display_name":"Simulated annealing","score":0.5087000131607056},{"id":"https://openalex.org/keywords/unavailability","display_name":"Unavailability","score":0.5031999945640564},{"id":"https://openalex.org/keywords/speedup","display_name":"Speedup","score":0.4424000084400177},{"id":"https://openalex.org/keywords/transformation-matrix","display_name":"Transformation matrix","score":0.43880000710487366},{"id":"https://openalex.org/keywords/equivalence","display_name":"Equivalence (formal languages)","score":0.4237000048160553},{"id":"https://openalex.org/keywords/matrix","display_name":"Matrix (chemical analysis)","score":0.3919999897480011}],"concepts":[{"id":"https://openalex.org/C190502265","wikidata":"https://www.wikidata.org/wiki/Q17069496","display_name":"MNIST database","level":3,"score":0.8521999716758728},{"id":"https://openalex.org/C11252640","wikidata":"https://www.wikidata.org/wiki/Q243723","display_name":"Monomial","level":2,"score":0.7940000295639038},{"id":"https://openalex.org/C186633575","wikidata":"https://www.wikidata.org/wiki/Q845060","display_name":"Maxima and minima","level":2,"score":0.769599974155426},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.5831000208854675},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5634999871253967},{"id":"https://openalex.org/C126980161","wikidata":"https://www.wikidata.org/wiki/Q863783","display_name":"Simulated annealing","level":2,"score":0.5087000131607056},{"id":"https://openalex.org/C2780505938","wikidata":"https://www.wikidata.org/wiki/Q17093282","display_name":"Unavailability","level":2,"score":0.5031999945640564},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.4968000054359436},{"id":"https://openalex.org/C68339613","wikidata":"https://www.wikidata.org/wiki/Q1549489","display_name":"Speedup","level":2,"score":0.4424000084400177},{"id":"https://openalex.org/C165443888","wikidata":"https://www.wikidata.org/wiki/Q1482183","display_name":"Transformation matrix","level":3,"score":0.43880000710487366},{"id":"https://openalex.org/C2780069185","wikidata":"https://www.wikidata.org/wiki/Q7977945","display_name":"Equivalence (formal languages)","level":2,"score":0.4237000048160553},{"id":"https://openalex.org/C106487976","wikidata":"https://www.wikidata.org/wiki/Q685816","display_name":"Matrix (chemical analysis)","level":2,"score":0.3919999897480011},{"id":"https://openalex.org/C2777303404","wikidata":"https://www.wikidata.org/wiki/Q759757","display_name":"Convergence (economics)","level":2,"score":0.3912999927997589},{"id":"https://openalex.org/C126255220","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Mathematical optimization","level":1,"score":0.3749000132083893},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.366100013256073},{"id":"https://openalex.org/C153258448","wikidata":"https://www.wikidata.org/wiki/Q1199743","display_name":"Gradient descent","level":3,"score":0.3422999978065491},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.3409000039100647},{"id":"https://openalex.org/C14036430","wikidata":"https://www.wikidata.org/wiki/Q3736076","display_name":"Function (biology)","level":2,"score":0.3391999900341034},{"id":"https://openalex.org/C204241405","wikidata":"https://www.wikidata.org/wiki/Q461499","display_name":"Transformation (genetics)","level":3,"score":0.3296000063419342},{"id":"https://openalex.org/C96469262","wikidata":"https://www.wikidata.org/wiki/Q1324364","display_name":"Homogeneous space","level":2,"score":0.32899999618530273},{"id":"https://openalex.org/C186661526","wikidata":"https://www.wikidata.org/wiki/Q13647261","display_name":"Sine","level":2,"score":0.3287000060081482},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.31859999895095825},{"id":"https://openalex.org/C22019652","wikidata":"https://www.wikidata.org/wiki/Q331309","display_name":"Overfitting","level":3,"score":0.31690001487731934},{"id":"https://openalex.org/C95623464","wikidata":"https://www.wikidata.org/wiki/Q1096149","display_name":"Classifier (UML)","level":2,"score":0.2919999957084656},{"id":"https://openalex.org/C177148314","wikidata":"https://www.wikidata.org/wiki/Q170084","display_name":"Generalization","level":2,"score":0.289000004529953},{"id":"https://openalex.org/C155032097","wikidata":"https://www.wikidata.org/wiki/Q798503","display_name":"Backpropagation","level":3,"score":0.28349998593330383},{"id":"https://openalex.org/C2776562905","wikidata":"https://www.wikidata.org/wiki/Q306610","display_name":"Hypersphere","level":2,"score":0.27730000019073486},{"id":"https://openalex.org/C51632099","wikidata":"https://www.wikidata.org/wiki/Q3985153","display_name":"Training set","level":2,"score":0.2694000005722046},{"id":"https://openalex.org/C141934464","wikidata":"https://www.wikidata.org/wiki/Q3305386","display_name":"Local optimum","level":2,"score":0.25769999623298645},{"id":"https://openalex.org/C137836250","wikidata":"https://www.wikidata.org/wiki/Q984063","display_name":"Optimization problem","level":2,"score":0.25609999895095825},{"id":"https://openalex.org/C47702885","wikidata":"https://www.wikidata.org/wiki/Q5441227","display_name":"Feedforward neural network","level":3,"score":0.25209999084472656},{"id":"https://openalex.org/C38858127","wikidata":"https://www.wikidata.org/wiki/Q5441228","display_name":"Feed forward","level":2,"score":0.2500999867916107}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/apsipaasc65261.2025.11249160","is_oa":false,"landing_page_url":"https://doi.org/10.1109/apsipaasc65261.2025.11249160","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 Asia Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA ASC)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":4,"referenced_works":["https://openalex.org/W1061737","https://openalex.org/W1498436455","https://openalex.org/W1988720110","https://openalex.org/W7103756444"],"related_works":[],"abstract_inverted_index":{"Gradient-based":[0],"optimizers":[1,13],"are":[2],"crucial":[3],"for":[4,97,166,182],"the":[5,63,67,71,82,86,90,103,140,146],"success":[6],"of":[7,66,131,158],"deep":[8,186],"learning":[9,187],"methods;":[10],"however,":[11],"such":[12],"frequently":[14],"suffer":[15],"from":[16],"slow":[17],"convergence":[18],"and":[19,93,113,120,154,184],"can":[20],"become":[21],"trapped":[22],"in":[23],"suboptimal":[24],"local":[25],"minima":[26],"or":[27],"plateaus.":[28],"To":[29],"overcome":[30],"this":[31],"issue,":[32],"we":[33],"propose":[34],"a":[35,53,74,155,179],"novel":[36],"optimization":[37],"technique":[38],"called":[39],"Monomial":[40],"Equivalence":[41],"-":[42],"Interleaved":[43],"Training":[44],"(ME-IT)":[45],"to":[46,73,88,145],"augment":[47],"well-known":[48],"optimizers.":[49],"ME-IT":[50],"periodically":[51],"applies":[52],"monomial":[54],"matrix":[55],"transform-a":[56],"computationally":[57],"cheap,":[58],"loss-preserving":[59],"transformation":[60],"that":[61,107,173],"leverages":[62],"inherent":[64],"symmetries":[65,177],"network.":[68],"This":[69,170],"\u201cre-locates\u201d":[70],"model":[72,147],"functionally":[75],"identical":[76],"but":[77],"parametrically":[78],"different":[79],"location":[80],"on":[81,102,139],"loss":[83,91,119],"level-set,":[84],"enabling":[85],"network":[87,176],"explore":[89],"surface":[92],"find":[94],"favorable":[95],"points":[96],"continued":[98],"gradient":[99],"descent.":[100],"Experiments":[101],"MNIST":[104,141],"dataset":[105,142],"demonstrate":[106],"ME-IT-applied":[108],"models":[109],"converge":[110],"significantly":[111],"faster":[112],"discover":[114],"final":[115],"solutions":[116],"with":[117,143,148,151],"lower":[118],"higher":[121],"accuracy":[122,129],"than":[123],"their":[124],"state-of-the-art":[125],"counterparts.":[126],"A":[127],"test":[128],"improvement":[130],"<tex":[132,160],"xmlns:mml=\"http://www.w3.org/1998/Math/MathML\"":[133,161],"xmlns:xlink=\"http://www.w3.org/1999/xlink\">$\\sim":[134],"0.3":[135],"\\%$</tex>":[136],"is":[137,164,178],"achieved":[138],"respect":[144],"Cosine":[149],"Annealing":[150],"Warm":[152],"Restarts,":[153],"training":[156,168],"speedup":[157],"around":[159],"xmlns:xlink=\"http://www.w3.org/1999/xlink\">$2":[162],"\\times$</tex>":[163],"reached":[165],"shorter":[167],"sessions.":[169],"work":[171],"establishes":[172],"strategically":[174],"leveraging":[175],"practical":[180],"tool":[181],"improving":[183],"accelerating":[185],"optimization.":[188]},"counts_by_year":[],"updated_date":"2026-06-26T08:34:08.712188","created_date":"2025-11-28T00:00:00"}