{"id":"https://openalex.org/W3143978893","doi":null,"title":"Branch-and-Pruning Optimization Towards Global Optimality in Deep Learning.","display_name":"Branch-and-Pruning Optimization Towards Global Optimality in Deep Learning.","publication_year":2021,"publication_date":"2021-04-05","ids":{"openalex":"https://openalex.org/W3143978893","mag":"3143978893"},"language":"en","primary_location":{"id":"mag:3143978893","doi":null,"is_oa":true,"landing_page_url":"http://export.arxiv.org/pdf/2104.01730","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":"http://export.arxiv.org/pdf/2104.01730","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5101520031","display_name":"Yuanwei Wu","orcid":"https://orcid.org/0000-0002-0385-8016","observed_orcids":["https://orcid.org/0000-0002-0385-8016"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yuanwei Wu","raw_affiliation_strings":[],"raw_orcid":"https://orcid.org/0000-0002-0385-8016","affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5048506332","display_name":"Ziming Zhang","orcid":"https://orcid.org/0000-0002-0921-8298","observed_orcids":["https://orcid.org/0000-0002-0921-8298"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Ziming Zhang","raw_affiliation_strings":[],"raw_orcid":"https://orcid.org/0000-0002-0921-8298","affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5100399485","display_name":"Guanghui Wang","orcid":"https://orcid.org/0000-0003-3182-104X","observed_orcids":["https://orcid.org/0000-0003-3182-104X"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Guanghui Wang","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":2,"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":0.9976000189781189,"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":0.9976000189781189,"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":0.0010000000474974513,"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/T10963","display_name":"Advanced Optimization Algorithms Research","score":0.00039999998989515007,"subfield":{"id":"https://openalex.org/subfields/2612","display_name":"Numerical Analysis"},"field":{"id":"https://openalex.org/fields/26","display_name":"Mathematics"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/global-optimality","display_name":"global optimality","score":0.8109999895095825},{"id":"https://openalex.org/keywords/deep-learning-optimization","display_name":"deep learning optimization","score":0.859000027179718},{"id":"https://openalex.org/keywords/adaptive-step-size","display_name":"adaptive step size","score":0.9150000214576721},{"id":"https://openalex.org/keywords/lipschitz-continuity","display_name":"Lipschitz continuity","score":0.9869999885559082},{"id":"https://openalex.org/keywords/gradient-based-optimization","display_name":"gradient-based optimization","score":0.6460000276565552},{"id":"https://openalex.org/keywords/optimization-algorithms","display_name":"optimization algorithms","score":0.382999986410141},{"id":"https://openalex.org/keywords/object-recognition","display_name":"object recognition","score":0.722000002861023},{"id":"https://openalex.org/keywords/object-detection","display_name":"object detection","score":0.9940000176429749},{"id":"https://openalex.org/keywords/image-segmentation","display_name":"image segmentation","score":0.734000027179718}],"concepts":[{"id":"https://openalex.org/C108010975","wikidata":"https://www.wikidata.org/wiki/Q500094","display_name":"Pruning","level":2,"score":0.8343935608863831},{"id":"https://openalex.org/C2778770139","wikidata":"https://www.wikidata.org/wiki/Q1966904","display_name":"Solver","level":2,"score":0.7530772686004639},{"id":"https://openalex.org/C22324862","wikidata":"https://www.wikidata.org/wiki/Q652707","display_name":"Lipschitz continuity","level":2,"score":0.6727901697158813},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6717244386672974},{"id":"https://openalex.org/C2776760102","wikidata":"https://www.wikidata.org/wiki/Q5139990","display_name":"Code (set theory)","level":3,"score":0.5673397779464722},{"id":"https://openalex.org/C164752517","wikidata":"https://www.wikidata.org/wiki/Q5570875","display_name":"Global optimization","level":2,"score":0.5596999526023865},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.5560231804847717},{"id":"https://openalex.org/C89600930","wikidata":"https://www.wikidata.org/wiki/Q1423946","display_name":"Segmentation","level":2,"score":0.5400034189224243},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5147245526313782},{"id":"https://openalex.org/C126255220","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Mathematical optimization","level":1,"score":0.49854350090026855},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.38475096225738525},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.2410047948360443},{"id":"https://openalex.org/C177264268","wikidata":"https://www.wikidata.org/wiki/Q1514741","display_name":"Set (abstract data type)","level":2,"score":0.0},{"id":"https://openalex.org/C6557445","wikidata":"https://www.wikidata.org/wiki/Q173113","display_name":"Agronomy","level":1,"score":0.0},{"id":"https://openalex.org/C86803240","wikidata":"https://www.wikidata.org/wiki/Q420","display_name":"Biology","level":0,"score":0.0},{"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/C199360897","wikidata":"https://www.wikidata.org/wiki/Q9143","display_name":"Programming language","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"mag:3143978893","doi":null,"is_oa":true,"landing_page_url":"http://export.arxiv.org/pdf/2104.01730","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:3143978893","doi":null,"is_oa":true,"landing_page_url":"http://export.arxiv.org/pdf/2104.01730","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":64,"referenced_works":["https://openalex.org/W6908809","https://openalex.org/W43577856","https://openalex.org/W104184427","https://openalex.org/W196871588","https://openalex.org/W639708223","https://openalex.org/W1522301498","https://openalex.org/W1686810756","https://openalex.org/W1845277745","https://openalex.org/W1899249567","https://openalex.org/W1903029394","https://openalex.org/W1963882359","https://openalex.org/W1980287119","https://openalex.org/W2088049833","https://openalex.org/W2091364465","https://openalex.org/W2102605133","https://openalex.org/W2108598243","https://openalex.org/W2112796928","https://openalex.org/W2117539524","https://openalex.org/W2130942839","https://openalex.org/W2146502635","https://openalex.org/W2146989110","https://openalex.org/W2160815625","https://openalex.org/W2194775991","https://openalex.org/W2399994860","https://openalex.org/W2404068508","https://openalex.org/W2436219157","https://openalex.org/W2474090883","https://openalex.org/W2523246573","https://openalex.org/W2591714514","https://openalex.org/W2597082880","https://openalex.org/W2605372163","https://openalex.org/W2608609325","https://openalex.org/W2617427826","https://openalex.org/W2662617432","https://openalex.org/W2736030546","https://openalex.org/W2746420172","https://openalex.org/W2793333878","https://openalex.org/W2803955134","https://openalex.org/W2806852391","https://openalex.org/W2810898455","https://openalex.org/W2894604724","https://openalex.org/W2899748887","https://openalex.org/W2900103278","https://openalex.org/W2900726521","https://openalex.org/W2900959181","https://openalex.org/W2908046483","https://openalex.org/W2912811302","https://openalex.org/W2950220847","https://openalex.org/W2962758212","https://openalex.org/W2962933129","https://openalex.org/W2963276946","https://openalex.org/W2963341956","https://openalex.org/W2963446712","https://openalex.org/W2963519230","https://openalex.org/W2963904364","https://openalex.org/W2963959597","https://openalex.org/W2964137095","https://openalex.org/W2964204240","https://openalex.org/W2991779855","https://openalex.org/W2995504922","https://openalex.org/W2996603555","https://openalex.org/W3010402260","https://openalex.org/W3035303608","https://openalex.org/W3093329015"],"related_works":["https://openalex.org/W2963276946","https://openalex.org/W3209032295","https://openalex.org/W3034400851","https://openalex.org/W2936762371","https://openalex.org/W2965794435","https://openalex.org/W7011775051","https://openalex.org/W3181808788","https://openalex.org/W4400373708","https://openalex.org/W6884991716","https://openalex.org/W3133231226"],"abstract_inverted_index":{"It":[0],"has":[1],"been":[2,24],"attracting":[3],"more":[4,6],"and":[5,51,67,130,141,150],"attention":[7],"to":[8,27],"understand":[9],"the":[10,60,75,79,83,95,110,144],"global":[11,31,96,111],"optimality":[12,112],"in":[13,65,143],"deep":[14,46],"learning":[15],"(DL)":[16],"recently.":[17],"However,":[18],"conventional":[19,133],"DL":[20,125,134],"solvers,":[21],"have":[22],"not":[23],"developed":[25],"intentionally":[26],"seek":[28],"for":[29,78,124],"such":[30,103,136],"optimality.":[32,97],"In":[33],"this":[34],"paper,":[35],"we":[36,107],"propose":[37],"a":[38,69,104],"novel":[39],"approximation":[40],"algorithm,":[41],"{\\em":[42],"BPGrad},":[43],"towards":[44],"optimizing":[45],"models":[47],"globally":[48],"via":[49],"branch":[50],"pruning.":[52],"The":[53,152],"proposed":[54,127],"BPGrad":[55,123],"algorithm":[56],"is":[57,126,154],"based":[58,121],"on":[59,122],"assumption":[61],"of":[62,85,146],"Lipschitz":[63],"continuity":[64],"DL,":[66],"as":[68,128,137],"result,":[70],"it":[71,131],"can":[72,93,108],"adaptively":[73],"determine":[74],"step":[76],"size":[77],"current":[80],"gradient":[81],"given":[82],"history":[84],"previous":[86],"updates,":[87],"wherein":[88],"theoretically":[89],"no":[90],"smaller":[91],"steps":[92],"achieve":[94],"We":[98],"prove":[99],"that,":[100],"by":[101],"repeating":[102],"branch-and-pruning":[105],"procedure,":[106],"locate":[109],"within":[113],"finite":[114],"iterations.":[115],"Empirically":[116],"an":[117],"efficient":[118],"adaptive":[119],"solver":[120],"well,":[129],"outperforms":[132],"solvers":[135],"Adagrad,":[138],"Adadelta,":[139],"RMSProp,":[140],"Adam":[142],"tasks":[145],"object":[147],"recognition,":[148],"detection,":[149],"segmentation.":[151],"code":[153],"available":[155],"at":[156],"\\url{this":[157],"https":[158],"URL}.":[159]},"counts_by_year":[{"year":2021,"cited_by_count":2}],"updated_date":"2026-10-16T09:59:22.235173","created_date":"2021-04-13T00:00:00"}