{"id":"https://openalex.org/W3137422635","doi":null,"title":"Train Deep Neural Networks in 40-D Subspaces.","display_name":"Train Deep Neural Networks in 40-D Subspaces.","publication_year":2021,"publication_date":"2021-03-20","ids":{"openalex":"https://openalex.org/W3137422635","mag":"3137422635"},"language":"en","primary_location":{"id":"mag:3137422635","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/pdf/2103.11154","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":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"arXiv (Cornell University)","raw_type":null},"volume":null,"type":"preprint","is_traditional":true,"indexed_in":[],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2103.11154","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5100455256","display_name":"Tao Li","orcid":"https://orcid.org/0000-0001-8966-7239","observed_orcids":["https://orcid.org/0000-0001-8966-7239"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Tao Li","raw_affiliation_strings":[],"raw_orcid":"https://orcid.org/0000-0001-8966-7239","affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5026903306","display_name":"Lei \u78ca Tan \u8c08","orcid":"https://orcid.org/0000-0001-6215-9242","observed_orcids":["https://orcid.org/0000-0001-6215-9242"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Lei \u78ca Tan \u8c08","raw_affiliation_strings":[],"raw_orcid":"https://orcid.org/0000-0001-6215-9242","affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5021602083","display_name":"Qinghua Tao","orcid":"https://orcid.org/0000-0001-9705-7748","observed_orcids":["https://orcid.org/0000-0001-9705-7748","https://orcid.org/0000-0002-2929-1919"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Qinghua Tao","raw_affiliation_strings":[],"raw_orcid":"https://orcid.org/0000-0001-9705-7748","affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100610065","display_name":"Yipeng Liu","orcid":"https://orcid.org/0000-0003-2084-8781","observed_orcids":["https://orcid.org/0000-0003-2084-8781","https://orcid.org/0009-0008-0429-3011"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yipeng Liu","raw_affiliation_strings":[],"raw_orcid":"https://orcid.org/0000-0003-2084-8781","affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5026169938","display_name":"Xiaolin Huang","orcid":"https://orcid.org/0000-0003-4285-6520","observed_orcids":["https://orcid.org/0000-0003-4285-6520"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xiaolin Huang","raw_affiliation_strings":[],"raw_orcid":"https://orcid.org/0000-0003-4285-6520","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":1,"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/T11612","display_name":"Stochastic Gradient Optimization Techniques","score":0.9872999787330627,"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":0.9872999787330627,"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/T10320","display_name":"Neural Networks and Applications","score":0.00839999970048666,"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":0.00139999995008111,"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/training-trajectories","display_name":"training trajectories","score":0.6859999895095825},{"id":"https://openalex.org/keywords/quasi-newton-methods","display_name":"quasi-Newton methods","score":0.8659999966621399},{"id":"https://openalex.org/keywords/second-order-optimization","display_name":"second-order optimization","score":0.9160000085830688},{"id":"https://openalex.org/keywords/deep-neural-network-training","display_name":"deep neural network training","score":0.7839999794960022},{"id":"https://openalex.org/keywords/parameter-space-reduction","display_name":"parameter space reduction","score":0.40299999713897705}],"concepts":[{"id":"https://openalex.org/C70518039","wikidata":"https://www.wikidata.org/wiki/Q16000077","display_name":"Dimensionality reduction","level":2,"score":0.7979511618614197},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.7869901657104492},{"id":"https://openalex.org/C111030470","wikidata":"https://www.wikidata.org/wiki/Q1430460","display_name":"Curse of dimensionality","level":2,"score":0.773373544216156},{"id":"https://openalex.org/C32834561","wikidata":"https://www.wikidata.org/wiki/Q660730","display_name":"Subspace topology","level":2,"score":0.7530122995376587},{"id":"https://openalex.org/C12362212","wikidata":"https://www.wikidata.org/wiki/Q728435","display_name":"Linear subspace","level":2,"score":0.7166738510131836},{"id":"https://openalex.org/C33676613","wikidata":"https://www.wikidata.org/wiki/Q13415176","display_name":"Dimension (graph theory)","level":2,"score":0.6578050851821899},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6172159314155579},{"id":"https://openalex.org/C111335779","wikidata":"https://www.wikidata.org/wiki/Q3454686","display_name":"Reduction (mathematics)","level":2,"score":0.5222744941711426},{"id":"https://openalex.org/C182365436","wikidata":"https://www.wikidata.org/wiki/Q50701","display_name":"Variable (mathematics)","level":2,"score":0.5031620860099792},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4987068176269531},{"id":"https://openalex.org/C2778572836","wikidata":"https://www.wikidata.org/wiki/Q380933","display_name":"Space (punctuation)","level":2,"score":0.44157329201698303},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.4232117235660553},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.2555520832538605},{"id":"https://openalex.org/C134306372","wikidata":"https://www.wikidata.org/wiki/Q7754","display_name":"Mathematical analysis","level":1,"score":0.0},{"id":"https://openalex.org/C202444582","wikidata":"https://www.wikidata.org/wiki/Q837863","display_name":"Pure mathematics","level":1,"score":0.0},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.0},{"id":"https://openalex.org/C2524010","wikidata":"https://www.wikidata.org/wiki/Q8087","display_name":"Geometry","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"mag:3137422635","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/pdf/2103.11154","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":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"arXiv (Cornell University)","raw_type":null}],"best_oa_location":{"id":"mag:3137422635","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/pdf/2103.11154","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":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"arXiv (Cornell University)","raw_type":null},"sustainable_development_goals":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":51,"referenced_works":["https://openalex.org/W587794757","https://openalex.org/W992687842","https://openalex.org/W1491468723","https://openalex.org/W1518461880","https://openalex.org/W1686810756","https://openalex.org/W1968157422","https://openalex.org/W1969811075","https://openalex.org/W2039708501","https://openalex.org/W2051434435","https://openalex.org/W2058839679","https://openalex.org/W2103111465","https://openalex.org/W2108598243","https://openalex.org/W2112796928","https://openalex.org/W2145680191","https://openalex.org/W2146502635","https://openalex.org/W2194775991","https://openalex.org/W2274287116","https://openalex.org/W2285660444","https://openalex.org/W2523246573","https://openalex.org/W2531409750","https://openalex.org/W2601450892","https://openalex.org/W2612445135","https://openalex.org/W2625567132","https://openalex.org/W2736941579","https://openalex.org/W2766839578","https://openalex.org/W2785533664","https://openalex.org/W2803187616","https://openalex.org/W2805003733","https://openalex.org/W2809090039","https://openalex.org/W2899748887","https://openalex.org/W2904243021","https://openalex.org/W2912322140","https://openalex.org/W2919115771","https://openalex.org/W2925181975","https://openalex.org/W2946948417","https://openalex.org/W2950967261","https://openalex.org/W2951037516","https://openalex.org/W2962933129","https://openalex.org/W2963420686","https://openalex.org/W2963446712","https://openalex.org/W2963697299","https://openalex.org/W2963794891","https://openalex.org/W2964045956","https://openalex.org/W2964121744","https://openalex.org/W2994081359","https://openalex.org/W3098238825","https://openalex.org/W3098987381","https://openalex.org/W3099571327","https://openalex.org/W3109493679","https://openalex.org/W3118608800","https://openalex.org/W4394670654"],"related_works":["https://openalex.org/W3193712339","https://openalex.org/W4282939141","https://openalex.org/W4312048098","https://openalex.org/W2904689686","https://openalex.org/W3177918757","https://openalex.org/W6941322927","https://openalex.org/W7221016620","https://openalex.org/W7214731359","https://openalex.org/W2766400409","https://openalex.org/W2083606636"],"abstract_inverted_index":{"Although":[0],"there":[1,51],"are":[2,52],"massive":[3],"parameters":[4,82],"in":[5,14],"deep":[6],"neural":[7,84,98],"networks,":[8,99],"the":[9,25,39,80,91,111],"training":[10,26,113],"can":[11,106],"actually":[12],"proceed":[13],"a":[15,30,43,54,68],"rather":[16,78],"low-dimensional":[17,22],"space.":[18],"By":[19],"investigating":[20],"such":[21],"properties":[23],"of":[24,46,83,119],"trajectory,":[27],"we":[28,66],"propose":[29],"Dynamic":[31],"Linear":[32],"Dimensionality":[33],"Reduction":[34],"(DLDR),":[35],"which":[36],"dramatically":[37],"reduces":[38],"parameter":[40],"space":[41],"to":[42,57,71],"variable":[44],"subspace":[45],"significantly":[47],"lower":[48],"dimension.":[49],"Since":[50],"only":[53,102],"few":[55],"variables":[56,74],"optimize,":[58],"second-order":[59],"methods":[60],"become":[61],"applicable.":[62],"Following":[63],"this":[64],"idea,":[65],"develop":[67],"quasi-Newton-based":[69],"algorithm":[70],"train":[72],"these":[73],"obtained":[75],"by":[76],"DLDR,":[77],"than":[79],"original":[81],"networks.":[85],"The":[86],"experimental":[87],"results":[88],"strongly":[89],"support":[90],"dimensionality":[92],"reduction":[93],"performance:":[94],"for":[95],"many":[96],"standard":[97],"optimizing":[100],"over":[101,114],"40":[103],"variables,":[104],"one":[105],"achieve":[107],"comparable":[108],"performance":[109],"against":[110],"regular":[112],"thousands":[115],"or":[116],"even":[117],"millions":[118],"parameters.":[120]},"counts_by_year":[{"year":2022,"cited_by_count":1}],"updated_date":"2026-11-03T10:50:19.383937","created_date":"2021-03-29T00:00:00"}