{"id":"https://openalex.org/W7161250662","doi":"https://doi.org/10.48550/arxiv.2605.14764","title":"Compositional Sparsity as an Inductive Bias for Neural Architecture Design","display_name":"Compositional Sparsity as an Inductive Bias for Neural Architecture Design","publication_year":2026,"publication_date":"2026-05-14","ids":{"openalex":"https://openalex.org/W7161250662","doi":"https://doi.org/10.48550/arxiv.2605.14764"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2605.14764","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.14764","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"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":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2605.14764","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5136217483","display_name":"Hongyu Lin","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Lin, Hongyu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5048066583","display_name":"Antonio Briola","orcid":"https://orcid.org/0000-0002-4391-0157"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Briola, Antonio","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5079237967","display_name":"Yuanrong Wang","orcid":"https://orcid.org/0000-0001-6708-4569"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Yuanrong","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5050674002","display_name":"Tomaso Aste","orcid":"https://orcid.org/0000-0002-4219-0215"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Aste, Tomaso","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":0,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"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/T11273","display_name":"Advanced Graph Neural Networks","score":0.7865999937057495,"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/T11273","display_name":"Advanced Graph Neural Networks","score":0.7865999937057495,"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/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.035100001841783524,"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/T12808","display_name":"Ferroelectric and Negative Capacitance Devices","score":0.020400000736117363,"subfield":{"id":"https://openalex.org/subfields/2208","display_name":"Electrical and Electronic Engineering"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/curse-of-dimensionality","display_name":"Curse of dimensionality","score":0.7735000252723694},{"id":"https://openalex.org/keywords/hyperparameter","display_name":"Hyperparameter","score":0.6265000104904175},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.6067000031471252},{"id":"https://openalex.org/keywords/prior-probability","display_name":"Prior probability","score":0.5587999820709229},{"id":"https://openalex.org/keywords/dependency","display_name":"Dependency (UML)","score":0.5210000276565552},{"id":"https://openalex.org/keywords/abstraction","display_name":"Abstraction","score":0.5181999802589417},{"id":"https://openalex.org/keywords/pipeline","display_name":"Pipeline (software)","score":0.4415000081062317},{"id":"https://openalex.org/keywords/suite","display_name":"Suite","score":0.4381999969482422},{"id":"https://openalex.org/keywords/variance","display_name":"Variance (accounting)","score":0.3862000107765198}],"concepts":[{"id":"https://openalex.org/C111030470","wikidata":"https://www.wikidata.org/wiki/Q1430460","display_name":"Curse of dimensionality","level":2,"score":0.7735000252723694},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6646999716758728},{"id":"https://openalex.org/C8642999","wikidata":"https://www.wikidata.org/wiki/Q4171168","display_name":"Hyperparameter","level":2,"score":0.6265000104904175},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.6152999997138977},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.6067000031471252},{"id":"https://openalex.org/C177769412","wikidata":"https://www.wikidata.org/wiki/Q278090","display_name":"Prior probability","level":3,"score":0.5587999820709229},{"id":"https://openalex.org/C19768560","wikidata":"https://www.wikidata.org/wiki/Q320727","display_name":"Dependency (UML)","level":2,"score":0.5210000276565552},{"id":"https://openalex.org/C124304363","wikidata":"https://www.wikidata.org/wiki/Q673661","display_name":"Abstraction","level":2,"score":0.5181999802589417},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.4675999879837036},{"id":"https://openalex.org/C43521106","wikidata":"https://www.wikidata.org/wiki/Q2165493","display_name":"Pipeline (software)","level":2,"score":0.4415000081062317},{"id":"https://openalex.org/C79581498","wikidata":"https://www.wikidata.org/wiki/Q1367530","display_name":"Suite","level":2,"score":0.4381999969482422},{"id":"https://openalex.org/C196083921","wikidata":"https://www.wikidata.org/wiki/Q7915758","display_name":"Variance (accounting)","level":2,"score":0.3862000107765198},{"id":"https://openalex.org/C182365436","wikidata":"https://www.wikidata.org/wiki/Q50701","display_name":"Variable (mathematics)","level":2,"score":0.3682999908924103},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.3677000105381012},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.3621000051498413},{"id":"https://openalex.org/C2984842247","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep neural networks","level":3,"score":0.3555000126361847},{"id":"https://openalex.org/C83546350","wikidata":"https://www.wikidata.org/wiki/Q1139051","display_name":"Regression","level":2,"score":0.3375999927520752},{"id":"https://openalex.org/C21200559","wikidata":"https://www.wikidata.org/wiki/Q7451068","display_name":"Sensitivity (control systems)","level":2,"score":0.3262999951839447},{"id":"https://openalex.org/C2779010991","wikidata":"https://www.wikidata.org/wiki/Q2720909","display_name":"Artifact (error)","level":2,"score":0.3070000112056732},{"id":"https://openalex.org/C199845137","wikidata":"https://www.wikidata.org/wiki/Q145490","display_name":"Network topology","level":2,"score":0.3068999946117401},{"id":"https://openalex.org/C25343380","wikidata":"https://www.wikidata.org/wiki/Q277521","display_name":"Relation (database)","level":2,"score":0.2939999997615814},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.2754000127315521},{"id":"https://openalex.org/C17137986","wikidata":"https://www.wikidata.org/wiki/Q215067","display_name":"Orthogonality","level":2,"score":0.2680000066757202},{"id":"https://openalex.org/C197352929","wikidata":"https://www.wikidata.org/wiki/Q1074074","display_name":"Inductive bias","level":4,"score":0.2639999985694885},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.2612000107765198},{"id":"https://openalex.org/C10485038","wikidata":"https://www.wikidata.org/wiki/Q48996162","display_name":"Hyperparameter optimization","level":3,"score":0.2547000050544739}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2605.14764","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.14764","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"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":null,"license_id":null,"version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2605.14764","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.14764","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"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":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Identifying":[0],"the":[1,12,73,83,125],"structural":[2],"priors":[3],"that":[4,27],"enable":[5],"Deep":[6],"Neural":[7,68],"Networks":[8,55,69],"(DNNs)":[9],"to":[10,167],"overcome":[11],"curse":[13],"of":[14,104,145],"dimensionality":[15,139],"is":[16,31],"a":[17,142],"fundamental":[18],"challenge":[19],"in":[20,94,132],"machine":[21],"learning":[22,30],"theory.":[23],"Existing":[24],"literature":[25],"suggests":[26],"effective":[28],"high-dimensional":[29],"driven":[32],"by":[33],"compositional":[34,127],"sparsity,":[35],"where":[36,134],"target":[37],"functions":[38],"decompose":[39],"into":[40,76],"constituents":[41],"supported":[42],"on":[43],"low-dimensional":[44],"variable":[45],"subsets.":[46],"To":[47],"investigate":[48],"this":[49,87],"hypothesis,":[50],"we":[51],"combine":[52],"Information":[53],"Filtering":[54],"(IFNs),":[56],"which":[57,71,95],"extract":[58],"sparse":[59,78,121],"dependency":[60],"structures":[61],"via":[62],"constrained":[63],"information":[64],"maximisation,":[65],"with":[66,119],"Homological":[67],"(HNNs),":[70],"map":[72],"inferred":[74],"topology":[75],"fixed-wiring":[77],"neural":[79],"graphs.":[80],"We":[81],"formalise":[82],"design":[84],"principles":[85],"underlying":[86,126],"construction":[88],"and":[89,110,129,163],"present":[90],"an":[91],"interpretable":[92],"pipeline":[93],"abstraction":[96],"emerges":[97],"through":[98],"hierarchical":[99],"composition.":[100],"HNNs":[101,123,148],"are":[102],"orders":[103],"magnitude":[105],"sparser":[106],"than":[107],"standard":[108],"DNNs":[109],"require":[111],"only":[112],"minimal":[113],"hyperparameter":[114],"tuning.":[115],"On":[116],"synthetic":[117],"tasks":[118],"known":[120],"hierarchies,":[122],"recover":[124],"structure":[128],"remain":[130],"stable":[131],"regimes":[133],"dense":[135,153],"alternatives":[136],"degrade":[137],"as":[138],"increases.":[140],"Across":[141],"broad":[143],"suite":[144],"real-world":[146],"datasets,":[147],"consistently":[149],"match":[150],"or":[151],"outperform":[152],"baselines":[154],"while":[155],"using":[156],"far":[157],"fewer":[158],"parameters,":[159],"exhibiting":[160],"lower":[161],"variance":[162],"showing":[164],"reduced":[165],"sensitivity":[166],"hyperparameters.":[168]},"counts_by_year":[],"updated_date":"2026-07-01T06:00:48.157686","created_date":"2026-05-16T00:00:00"}