{"id":"https://openalex.org/W7214923139","doi":"https://doi.org/10.48550/arxiv.2609.36084","title":"Measuring trainable degrees of freedom in materials graph neural networks: a random-subspace intrinsic dimension analysis","display_name":"Measuring trainable degrees of freedom in materials graph neural networks: a random-subspace intrinsic dimension analysis","publication_year":2026,"publication_date":"2026-09-28","ids":{"openalex":"https://openalex.org/W7214923139","doi":"https://doi.org/10.48550/arxiv.2609.36084"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2609.36084","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2609.36084","pdf_url":"https://arxiv.org/pdf/2609.36084","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":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"volume":null,"type":"preprint","is_traditional":true,"indexed_in":["arxiv","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2609.36084","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5146398972","display_name":"Shehroz Ahmad Shoaib","orcid":"https://orcid.org/0009-0008-5362-2184","observed_orcids":["https://orcid.org/0009-0008-5362-2184"]},"institutions":[{"id":"https://openalex.org/I134085113","display_name":"King Fahd University of Petroleum and Minerals","ror":"https://ror.org/03yez3163","country_code":"SA","type":"education","lineage":["https://openalex.org/I134085113"]}],"countries":["SA"],"is_corresponding":false,"raw_author_name":"Shoaib, Shehroz Ahmad","raw_affiliation_strings":["Department of Electrical Engineering , King Fahd University of Petroleum and Minerals , Dhahran , Saudi Arabia"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Department of Electrical Engineering , King Fahd University of Petroleum and Minerals , Dhahran , Saudi Arabia","institution_ids":["https://openalex.org/I134085113"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5153672119","display_name":"Kangming Li","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I71920554","display_name":"King Abdullah University of Science and Technology","ror":"https://ror.org/01q3tbs38","country_code":"SA","type":"education","lineage":["https://openalex.org/I71920554"]}],"countries":["SA"],"is_corresponding":false,"raw_author_name":"Li, Kangming","raw_affiliation_strings":["Physical Science and Engineering Division , King Abdullah University of Science and Technology , Thuwal , Saudi Arabia"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Physical Science and Engineering Division , King Abdullah University of Science and Technology , Thuwal , Saudi Arabia","institution_ids":["https://openalex.org/I71920554"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"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/T11948","display_name":"Machine Learning in Materials Science","score":1.0,"subfield":{"id":"https://openalex.org/subfields/2505","display_name":"Materials Chemistry"},"field":{"id":"https://openalex.org/fields/25","display_name":"Materials Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},"topics":[{"id":"https://openalex.org/T11948","display_name":"Machine Learning in Materials Science","score":1.0,"subfield":{"id":"https://openalex.org/subfields/2505","display_name":"Materials Chemistry"},"field":{"id":"https://openalex.org/fields/25","display_name":"Materials Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T10002","display_name":"Advanced Chemical Physics Studies","score":0.0,"subfield":{"id":"https://openalex.org/subfields/3107","display_name":"Atomic and Molecular Physics, and Optics"},"field":{"id":"https://openalex.org/fields/31","display_name":"Physics and Astronomy"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T10211","display_name":"Computational Drug Discovery Methods","score":0.0,"subfield":{"id":"https://openalex.org/subfields/3002","display_name":"Drug Discovery"},"field":{"id":"https://openalex.org/fields/30","display_name":"Pharmacology, Toxicology and Pharmaceutics"},"domain":{"id":"https://openalex.org/domains/1","display_name":"Life Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/alignn","display_name":"ALIGNN","score":0.9980000257492065},{"id":"https://openalex.org/keywords/band-gap-prediction","display_name":"band gap prediction","score":0.7770000100135803},{"id":"https://openalex.org/keywords/formation-energy-prediction","display_name":"formation energy prediction","score":0.8379999995231628},{"id":"https://openalex.org/keywords/model-capacity","display_name":"model capacity","score":0.3580000102519989},{"id":"https://openalex.org/keywords/dataset-size","display_name":"dataset size","score":0.4749999940395355}],"concepts":[{"id":"https://openalex.org/C12362212","wikidata":"https://www.wikidata.org/wiki/Q728435","display_name":"Linear subspace","level":2,"score":0.7753000259399414},{"id":"https://openalex.org/C208081375","wikidata":"https://www.wikidata.org/wiki/Q274502","display_name":"Degrees of freedom (physics and chemistry)","level":2,"score":0.6621999740600586},{"id":"https://openalex.org/C33676613","wikidata":"https://www.wikidata.org/wiki/Q13415176","display_name":"Dimension (graph theory)","level":2,"score":0.5910000205039978},{"id":"https://openalex.org/C132525143","wikidata":"https://www.wikidata.org/wiki/Q141488","display_name":"Graph","level":2,"score":0.5403000116348267},{"id":"https://openalex.org/C2780009758","wikidata":"https://www.wikidata.org/wiki/Q6804172","display_name":"Measure (data warehouse)","level":2,"score":0.5192000269889832},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.5091999769210815},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.46810001134872437},{"id":"https://openalex.org/C2780841128","wikidata":"https://www.wikidata.org/wiki/Q5073781","display_name":"Characterization (materials science)","level":2,"score":0.4562000036239624},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.3953000009059906},{"id":"https://openalex.org/C83546350","wikidata":"https://www.wikidata.org/wiki/Q1139051","display_name":"Regression","level":2,"score":0.3935999870300293},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.3840000033378601},{"id":"https://openalex.org/C48921125","wikidata":"https://www.wikidata.org/wiki/Q10861030","display_name":"Linear regression","level":2,"score":0.375900000333786},{"id":"https://openalex.org/C12426560","wikidata":"https://www.wikidata.org/wiki/Q189569","display_name":"Basis (linear algebra)","level":2,"score":0.3734000027179718},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.36059999465942383},{"id":"https://openalex.org/C152877465","wikidata":"https://www.wikidata.org/wiki/Q208042","display_name":"Regression analysis","level":2,"score":0.3603000044822693},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.3278999924659729},{"id":"https://openalex.org/C28719098","wikidata":"https://www.wikidata.org/wiki/Q44946","display_name":"Point (geometry)","level":2,"score":0.3188000023365021},{"id":"https://openalex.org/C2777115002","wikidata":"https://www.wikidata.org/wiki/Q7168246","display_name":"Performance prediction","level":2,"score":0.31040000915527344},{"id":"https://openalex.org/C169191187","wikidata":"https://www.wikidata.org/wiki/Q1418353","display_name":"Multidimensional analysis","level":2,"score":0.3091000020503998},{"id":"https://openalex.org/C45804977","wikidata":"https://www.wikidata.org/wiki/Q7239673","display_name":"Predictive modelling","level":2,"score":0.30640000104904175},{"id":"https://openalex.org/C21036866","wikidata":"https://www.wikidata.org/wiki/Q181767","display_name":"Stress (linguistics)","level":2,"score":0.2874000072479248},{"id":"https://openalex.org/C21080849","wikidata":"https://www.wikidata.org/wiki/Q13611879","display_name":"Data point","level":2,"score":0.2547999918460846},{"id":"https://openalex.org/C163175372","wikidata":"https://www.wikidata.org/wiki/Q3339222","display_name":"Linear model","level":2,"score":0.2526000142097473}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2609.36084","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2609.36084","pdf_url":"https://arxiv.org/pdf/2609.36084","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":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"doi:10.48550/arxiv.2609.36084","doi":"https://doi.org/10.48550/arxiv.2609.36084","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2609.36084","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:2609.36084","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2609.36084","pdf_url":"https://arxiv.org/pdf/2609.36084","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":null,"license_id":null,"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/F4320322320","display_name":"King Abdullah University of Science and Technology","ror":"https://ror.org/01q3tbs38"}],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7214923139.pdf","grobid_xml":"https://content.openalex.org/works/W7214923139.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W4385755683","https://openalex.org/W4301180830","https://openalex.org/W7214531260","https://openalex.org/W4391994904","https://openalex.org/W7214545535","https://openalex.org/W4399552239","https://openalex.org/W3125542198","https://openalex.org/W4400438668","https://openalex.org/W7211960652","https://openalex.org/W4386510996"],"abstract_inverted_index":{"Final":[0],"predictive":[1],"accuracy":[2,82],"is":[3,122],"the":[4,84],"standard":[5],"basis":[6],"for":[7,176],"comparing":[8],"graph":[9],"neural":[10],"networks":[11],"(GNNs)":[12],"in":[13,54],"materials-property":[14],"prediction,":[15],"but":[16],"it":[17],"does":[18],"not":[19],"show":[20,115,130],"how":[21,65,177],"strongly":[22],"performance":[23,66,106],"depends":[24],"on":[25],"access":[26],"to":[27,88,125],"trainable":[28,70],"parameter-space":[29],"directions.":[30],"Here,":[31],"we":[32,48],"introduce":[33],"trainable-degree":[34],"dependence":[35],"as":[36,68,138],"a":[37,172],"complementary":[38],"characterization":[39],"of":[40,72],"materials":[41,178],"GNN":[42],"learning.":[43],"Using":[44],"random-subspace":[45],"intrinsic-dimension":[46],"analysis,":[47],"train":[49],"CGCNN,":[50],"ALIGNN,":[51],"and":[52,63,101,112,119,144],"DimeNet++":[53],"randomly":[55],"oriented":[56],"parameter":[57],"subspaces":[58,137],"across":[59],"six":[60],"prediction":[61,114,121],"tasks":[62],"measure":[64],"recovers":[67],"independent":[69],"degrees":[71],"freedom":[73],"are":[74,147],"restored.":[75],"The":[76],"resulting":[77],"recovery":[78],"curves":[79],"separate":[80],"endpoint":[81],"from":[83,107],"trainable-dimensional":[85],"demand":[86],"required":[87],"recover":[89,104],"it.":[90],"They":[91],"reveal":[92],"distinctions":[93],"that":[94,131,154],"final":[95],"errors":[96],"alone":[97],"miss:":[98],"metallic":[99],"classification":[100],"log-bulk-modulus":[102],"regression":[103],"near-reference":[105],"small":[108],"fractional":[109,136,155],"subspaces,":[110],"formation-energy":[111,143],"band-gap":[113,132],"stronger":[116],"architecture":[117],"dependence,":[118],"phonon":[120],"most":[123],"sensitive":[124],"dimensional":[126],"restriction.":[127],"Dataset-size":[128],"sweeps":[129],"models":[133],"require":[134],"larger":[135],"training":[139],"data":[140],"grows,":[141],"whereas":[142],"bulk-modulus":[145],"responses":[146],"more":[148],"stable.":[149],"A":[150],"width":[151],"sweep":[152],"shows":[153],"thresholds":[156],"can":[157],"remain":[158],"stable":[159],"while":[160],"absolute":[161],"threshold":[162],"dimensions":[163],"increase":[164],"with":[165],"model":[166],"size.":[167],"Random-subspace":[168],"analysis":[169],"therefore":[170],"provides":[171],"targeted":[173],"stress":[174],"test":[175],"GNNs":[179],"use":[180],"their":[181],"optimization":[182],"space.":[183]},"counts_by_year":[],"updated_date":"2026-10-28T10:33:24.094632","created_date":"2026-10-01T00:00:00"}