MLAAN: Scaling Supervised Local Learning with Multilaminar Leap Augmented Auxiliary Network Article Swipe

PDF

Yuming Zhang , Shouxin Zhang , Peizhe Wang , Feiyu Zhu , Dongzhi Guan , Junhao Su , Jiabin Liu , Changpeng Cai ·

YOU? · · 2025 · Open Access · · DOI: https://doi.org/10.1609/aaai.v39i21.34428

Deep neural networks (DNNs) typically employ an end-to-end (E2E) training paradigm which presents several challenges, including high GPU memory consumption, inefficiency, and difficulties in model parallelization during training. Recent research has sought to address these issues, with one promising approach being local learning. This method involves partitioning the backbone network into gradient-isolated modules and manually designing auxiliary networks to train these local modules. Existing methods often neglect the interaction of information between local modules, leading to myopic issues and a performance gap compared to E2E training. To address these limitations, we propose the Multilaminar Leap Augmented Auxiliary Network (MLAAN). Specifically, MLAAN comprises Multilaminar Local Modules (MLM) and Leap Augmented Modules (LAM). MLM captures both local and global features through independent and cascaded auxiliary networks, alleviating performance issues caused by insufficient global features. However, overly simplistic auxiliary networks can impede MLM's ability to capture global information. To address this, we further design LAM, an enhanced auxiliary network that uses the Exponential Moving Average (EMA) method to facilitate information exchange between local modules, thereby mitigating the shortsightedness resulting from inadequate interaction. The synergy between MLM and LAM has demonstrated excellent performance. Our experiments on the CIFAR-10, STL-10, SVHN, and ImageNet datasets show that MLAAN can be seamlessly integrated into existing local learning frameworks, significantly enhancing their performance and even surpassing end-to-end (E2E) training methods, while also reducing GPU memory consumption.

Related Topics

Computer Science

Artificial Intelligence

Mathematics

Geometry

Concepts

Scaling Computer science Artificial intelligence Mathematics Geometry

Metadata

Type: article
Language: en
Landing Page: https://doi.org/10.1609/aaai.v39i21.34428
PDF: https://ojs.aaai.org/index.php/AAAI/article/download/34428/36583
OA Status: diamond
Related Works: 10
OpenAlex ID: https://openalex.org/W4409363258

All OpenAlex metadata

Raw OpenAlex JSON

OpenAlex ID: https://openalex.org/W4409363258

Canonical identifier for this work in OpenAlex
DOI: https://doi.org/10.1609/aaai.v39i21.34428

Digital Object Identifier
Title: MLAAN: Scaling Supervised Local Learning with Multilaminar Leap Augmented Auxiliary Network

Work title
Type: article

OpenAlex work type
Language: en

Primary language
Publication year: 2025

Year of publication
Publication date: 2025-04-11

Full publication date if available
Authors: Yuming Zhang, Shouxin Zhang, Peizhe Wang, Feiyu Zhu, Dongzhi Guan, Junhao Su, Jiabin Liu, Changpeng Cai

List of authors in order
Landing page: https://doi.org/10.1609/aaai.v39i21.34428

Publisher landing page
PDF URL: https://ojs.aaai.org/index.php/AAAI/article/download/34428/36583

Direct link to full text PDF
Open access: Yes

Whether a free full text is available
OA status: diamond

Open access status per OpenAlex
OA URL: https://ojs.aaai.org/index.php/AAAI/article/download/34428/36583

Direct OA link when available
Concepts: Scaling, Computer science, Artificial intelligence, Mathematics, Geometry

Top concepts (fields/topics) attached by OpenAlex
Cited by: 0

Total citation count in OpenAlex
Related works (count): 10

Other works algorithmically related by OpenAlex

Full payload

id	https://openalex.org/W4409363258
doi	https://doi.org/10.1609/aaai.v39i21.34428
ids.doi	https://doi.org/10.1609/aaai.v39i21.34428
ids.openalex	https://openalex.org/W4409363258
fwci	0.0
type	article
title	MLAAN: Scaling Supervised Local Learning with Multilaminar Leap Augmented Auxiliary Network
biblio.issue	21
biblio.volume	39
biblio.last_page	22694
biblio.first_page	22686
topics[0].id	https://openalex.org/T10057
topics[0].field.id	https://openalex.org/fields/17
topics[0].field.display_name	Computer Science
topics[0].score	0.8105999827384949
topics[0].domain.id	https://openalex.org/domains/3
topics[0].domain.display_name	Physical Sciences
topics[0].subfield.id	https://openalex.org/subfields/1707
topics[0].subfield.display_name	Computer Vision and Pattern Recognition
topics[0].display_name	Face and Expression Recognition
topics[1].id	https://openalex.org/T11550
topics[1].field.id	https://openalex.org/fields/17
topics[1].field.display_name	Computer Science
topics[1].score	0.8004000186920166
topics[1].domain.id	https://openalex.org/domains/3
topics[1].domain.display_name	Physical Sciences
topics[1].subfield.id	https://openalex.org/subfields/1702
topics[1].subfield.display_name	Artificial Intelligence
topics[1].display_name	Text and Document Classification Technologies
is_xpac	False
apc_list
apc_paid
concepts[0].id	https://openalex.org/C99844830
concepts[0].level	2
concepts[0].score	0.6847906112670898
concepts[0].wikidata	https://www.wikidata.org/wiki/Q102441924
concepts[0].display_name	Scaling
concepts[1].id	https://openalex.org/C41008148
concepts[1].level	0
concepts[1].score	0.4667285680770874
concepts[1].wikidata	https://www.wikidata.org/wiki/Q21198
concepts[1].display_name	Computer science
concepts[2].id	https://openalex.org/C154945302
concepts[2].level	1
concepts[2].score	0.4364296793937683
concepts[2].wikidata	https://www.wikidata.org/wiki/Q11660
concepts[2].display_name	Artificial intelligence
concepts[3].id	https://openalex.org/C33923547
concepts[3].level	0
concepts[3].score	0.1617635190486908
concepts[3].wikidata	https://www.wikidata.org/wiki/Q395
concepts[3].display_name	Mathematics
concepts[4].id	https://openalex.org/C2524010
concepts[4].level	1
concepts[4].score	0.0
concepts[4].wikidata	https://www.wikidata.org/wiki/Q8087
concepts[4].display_name	Geometry
keywords[0].id	https://openalex.org/keywords/scaling
keywords[0].score	0.6847906112670898
keywords[0].display_name	Scaling
keywords[1].id	https://openalex.org/keywords/computer-science
keywords[1].score	0.4667285680770874
keywords[1].display_name	Computer science
keywords[2].id	https://openalex.org/keywords/artificial-intelligence
keywords[2].score	0.4364296793937683
keywords[2].display_name	Artificial intelligence
keywords[3].id	https://openalex.org/keywords/mathematics
keywords[3].score	0.1617635190486908
keywords[3].display_name	Mathematics
language	en
locations[0].id	doi:10.1609/aaai.v39i21.34428
locations[0].is_oa	True
locations[0].source.id	https://openalex.org/S4210191458
locations[0].source.issn	2159-5399, 2374-3468
locations[0].source.type	conference
locations[0].source.is_oa	True
locations[0].source.issn_l	2159-5399
locations[0].source.is_core	False
locations[0].source.is_in_doaj	False
locations[0].source.display_name	Proceedings of the AAAI Conference on Artificial Intelligence
locations[0].source.host_organization	https://openalex.org/P4310320058
locations[0].source.host_organization_name	Association for the Advancement of Artificial Intelligence
locations[0].source.host_organization_lineage	https://openalex.org/P4310320058
locations[0].source.host_organization_lineage_names	Association for the Advancement of Artificial Intelligence
locations[0].license
locations[0].pdf_url	https://ojs.aaai.org/index.php/AAAI/article/download/34428/36583
locations[0].version	publishedVersion
locations[0].raw_type	journal-article
locations[0].license_id
locations[0].is_accepted	True
locations[0].is_published	True
locations[0].raw_source_name	Proceedings of the AAAI Conference on Artificial Intelligence
locations[0].landing_page_url	https://doi.org/10.1609/aaai.v39i21.34428
indexed_in	crossref
authorships[0].author.id	https://openalex.org/A5101435876
authorships[0].author.orcid	https://orcid.org/0000-0002-8206-7457
authorships[0].author.display_name	Yuming Zhang
authorships[0].author_position	first
authorships[0].raw_author_name	Yuming Zhang
authorships[0].is_corresponding	False
authorships[1].author.id	https://openalex.org/A5050321672
authorships[1].author.orcid	https://orcid.org/0000-0002-5396-0393
authorships[1].author.display_name	Shouxin Zhang
authorships[1].author_position	middle
authorships[1].raw_author_name	Shouxin Zhang
authorships[1].is_corresponding	False
authorships[2].author.id	https://openalex.org/A5063742204
authorships[2].author.orcid	https://orcid.org/0000-0003-4298-2471
authorships[2].author.display_name	Peizhe Wang
authorships[2].author_position	middle
authorships[2].raw_author_name	Peizhe Wang
authorships[2].is_corresponding	False
authorships[3].author.id	https://openalex.org/A5025907387
authorships[3].author.orcid	https://orcid.org/0000-0002-9971-1667
authorships[3].author.display_name	Feiyu Zhu
authorships[3].author_position	middle
authorships[3].raw_author_name	Feiyu Zhu
authorships[3].is_corresponding	False
authorships[4].author.id	https://openalex.org/A5114189702
authorships[4].author.orcid
authorships[4].author.display_name	Dongzhi Guan
authorships[4].author_position	middle
authorships[4].raw_author_name	Dongzhi Guan
authorships[4].is_corresponding	False
authorships[5].author.id	https://openalex.org/A5109737760
authorships[5].author.orcid
authorships[5].author.display_name	Junhao Su
authorships[5].author_position	middle
authorships[5].raw_author_name	Junhao Su
authorships[5].is_corresponding	False
authorships[6].author.id	https://openalex.org/A5100659171
authorships[6].author.orcid	https://orcid.org/0000-0001-6914-8941
authorships[6].author.display_name	Jiabin Liu
authorships[6].author_position	middle
authorships[6].raw_author_name	Jiabin Liu
authorships[6].is_corresponding	False
authorships[7].author.id	https://openalex.org/A5102782423
authorships[7].author.orcid	https://orcid.org/0009-0007-1600-0784
authorships[7].author.display_name	Changpeng Cai
authorships[7].author_position	last
authorships[7].raw_author_name	Changpeng Cai
authorships[7].is_corresponding	False
has_content.pdf	True
has_content.grobid_xml	True
is_paratext	False
open_access.is_oa	True
open_access.oa_url	https://ojs.aaai.org/index.php/AAAI/article/download/34428/36583
open_access.oa_status	diamond
open_access.any_repository_has_fulltext	False
created_date	2025-10-10T00:00:00
display_name	MLAAN: Scaling Supervised Local Learning with Multilaminar Leap Augmented Auxiliary Network
has_fulltext	False
is_retracted	False
updated_date	2025-11-06T03:46:38.306776
primary_topic.id	https://openalex.org/T10057
primary_topic.field.id	https://openalex.org/fields/17
primary_topic.field.display_name	Computer Science
primary_topic.score	0.8105999827384949
primary_topic.domain.id	https://openalex.org/domains/3
primary_topic.domain.display_name	Physical Sciences
primary_topic.subfield.id	https://openalex.org/subfields/1707
primary_topic.subfield.display_name	Computer Vision and Pattern Recognition
primary_topic.display_name	Face and Expression Recognition
related_works	https://openalex.org/W4391375266, https://openalex.org/W2899084033, https://openalex.org/W2748952813, https://openalex.org/W2390279801, https://openalex.org/W4391913857, https://openalex.org/W2358668433, https://openalex.org/W4396701345, https://openalex.org/W2376932109, https://openalex.org/W2001405890, https://openalex.org/W4396696052
cited_by_count	0
locations_count	1
best_oa_location.id	doi:10.1609/aaai.v39i21.34428
best_oa_location.is_oa	True
best_oa_location.source.id	https://openalex.org/S4210191458
best_oa_location.source.issn	2159-5399, 2374-3468
best_oa_location.source.type	conference
best_oa_location.source.is_oa	True
best_oa_location.source.issn_l	2159-5399
best_oa_location.source.is_core	False
best_oa_location.source.is_in_doaj	False
best_oa_location.source.display_name	Proceedings of the AAAI Conference on Artificial Intelligence
best_oa_location.source.host_organization	https://openalex.org/P4310320058
best_oa_location.source.host_organization_name	Association for the Advancement of Artificial Intelligence
best_oa_location.source.host_organization_lineage	https://openalex.org/P4310320058
best_oa_location.source.host_organization_lineage_names	Association for the Advancement of Artificial Intelligence
best_oa_location.license
best_oa_location.pdf_url	https://ojs.aaai.org/index.php/AAAI/article/download/34428/36583
best_oa_location.version	publishedVersion
best_oa_location.raw_type	journal-article
best_oa_location.license_id
best_oa_location.is_accepted	True
best_oa_location.is_published	True
best_oa_location.raw_source_name	Proceedings of the AAAI Conference on Artificial Intelligence
best_oa_location.landing_page_url	https://doi.org/10.1609/aaai.v39i21.34428
primary_location.id	doi:10.1609/aaai.v39i21.34428
primary_location.is_oa	True
primary_location.source.id	https://openalex.org/S4210191458
primary_location.source.issn	2159-5399, 2374-3468
primary_location.source.type	conference
primary_location.source.is_oa	True
primary_location.source.issn_l	2159-5399
primary_location.source.is_core	False
primary_location.source.is_in_doaj	False
primary_location.source.display_name	Proceedings of the AAAI Conference on Artificial Intelligence
primary_location.source.host_organization	https://openalex.org/P4310320058
primary_location.source.host_organization_name	Association for the Advancement of Artificial Intelligence
primary_location.source.host_organization_lineage	https://openalex.org/P4310320058
primary_location.source.host_organization_lineage_names	Association for the Advancement of Artificial Intelligence
primary_location.license
primary_location.pdf_url	https://ojs.aaai.org/index.php/AAAI/article/download/34428/36583
primary_location.version	publishedVersion
primary_location.raw_type	journal-article
primary_location.license_id
primary_location.is_accepted	True
primary_location.is_published	True
primary_location.raw_source_name	Proceedings of the AAAI Conference on Artificial Intelligence
primary_location.landing_page_url	https://doi.org/10.1609/aaai.v39i21.34428
publication_date	2025-04-11
publication_year	2025
referenced_works_count	0
abstract_inverted_index.a	79
abstract_inverted_index.To	86, 145
abstract_inverted_index.an	6, 152
abstract_inverted_index.be	203
abstract_inverted_index.by	128
abstract_inverted_index.in	23
abstract_inverted_index.of	69
abstract_inverted_index.on	191
abstract_inverted_index.to	32, 58, 75, 83, 141, 164
abstract_inverted_index.we	90, 148
abstract_inverted_index.E2E	84
abstract_inverted_index.GPU	17, 225
abstract_inverted_index.LAM	184
abstract_inverted_index.MLM	111, 182
abstract_inverted_index.Our	189
abstract_inverted_index.The	179
abstract_inverted_index.and	21, 53, 78, 106, 115, 120, 183, 196, 215
abstract_inverted_index.can	137, 202
abstract_inverted_index.gap	81
abstract_inverted_index.has	30, 185
abstract_inverted_index.one	37
abstract_inverted_index.the	47, 67, 92, 158, 173, 192
abstract_inverted_index.Deep	0
abstract_inverted_index.LAM,	151
abstract_inverted_index.Leap	94, 107
abstract_inverted_index.This	43
abstract_inverted_index.also	223
abstract_inverted_index.both	113
abstract_inverted_index.even	216
abstract_inverted_index.from	176
abstract_inverted_index.high	16
abstract_inverted_index.into	50, 206
abstract_inverted_index.show	199
abstract_inverted_index.that	156, 200
abstract_inverted_index.uses	157
abstract_inverted_index.with	36
abstract_inverted_index.(E2E)	8, 219
abstract_inverted_index.(EMA)	162
abstract_inverted_index.(MLM)	105
abstract_inverted_index.Local	103
abstract_inverted_index.MLAAN	100, 201
abstract_inverted_index.MLM's	139
abstract_inverted_index.SVHN,	195
abstract_inverted_index.being	40
abstract_inverted_index.local	41, 61, 72, 114, 169, 208
abstract_inverted_index.model	24
abstract_inverted_index.often	65
abstract_inverted_index.their	213
abstract_inverted_index.these	34, 60, 88
abstract_inverted_index.this,	147
abstract_inverted_index.train	59
abstract_inverted_index.which	11
abstract_inverted_index.while	222
abstract_inverted_index.(DNNs)	3
abstract_inverted_index.(LAM).	110
abstract_inverted_index.Moving	160
abstract_inverted_index.Recent	28
abstract_inverted_index.caused	127
abstract_inverted_index.design	150
abstract_inverted_index.during	26
abstract_inverted_index.employ	5
abstract_inverted_index.global	116, 130, 143
abstract_inverted_index.impede	138
abstract_inverted_index.issues	77, 126
abstract_inverted_index.memory	18, 226
abstract_inverted_index.method	44, 163
abstract_inverted_index.myopic	76
abstract_inverted_index.neural	1
abstract_inverted_index.overly	133
abstract_inverted_index.sought	31
abstract_inverted_index.Average	161
abstract_inverted_index.Modules	104, 109
abstract_inverted_index.Network	97
abstract_inverted_index.STL-10,	194
abstract_inverted_index.ability	140
abstract_inverted_index.address	33, 87, 146
abstract_inverted_index.between	71, 168, 181
abstract_inverted_index.capture	142
abstract_inverted_index.further	149
abstract_inverted_index.issues,	35
abstract_inverted_index.leading	74
abstract_inverted_index.methods	64
abstract_inverted_index.modules	52
abstract_inverted_index.neglect	66
abstract_inverted_index.network	49, 155
abstract_inverted_index.propose	91
abstract_inverted_index.several	13
abstract_inverted_index.synergy	180
abstract_inverted_index.thereby	171
abstract_inverted_index.through	118
abstract_inverted_index.(MLAAN).	98
abstract_inverted_index.Existing	63
abstract_inverted_index.However,	132
abstract_inverted_index.ImageNet	197
abstract_inverted_index.approach	39
abstract_inverted_index.backbone	48
abstract_inverted_index.captures	112
abstract_inverted_index.cascaded	121
abstract_inverted_index.compared	82
abstract_inverted_index.datasets	198
abstract_inverted_index.enhanced	153
abstract_inverted_index.exchange	167
abstract_inverted_index.existing	207
abstract_inverted_index.features	117
abstract_inverted_index.involves	45
abstract_inverted_index.learning	209
abstract_inverted_index.manually	54
abstract_inverted_index.methods,	221
abstract_inverted_index.modules,	73, 170
abstract_inverted_index.modules.	62
abstract_inverted_index.networks	2, 57, 136
abstract_inverted_index.paradigm	10
abstract_inverted_index.presents	12
abstract_inverted_index.reducing	224
abstract_inverted_index.research	29
abstract_inverted_index.training	9, 220
abstract_inverted_index.Augmented	95, 108
abstract_inverted_index.Auxiliary	96
abstract_inverted_index.CIFAR-10,	193
abstract_inverted_index.auxiliary	56, 122, 135, 154
abstract_inverted_index.comprises	101
abstract_inverted_index.designing	55
abstract_inverted_index.enhancing	212
abstract_inverted_index.excellent	187
abstract_inverted_index.features.	131
abstract_inverted_index.including	15
abstract_inverted_index.learning.	42
abstract_inverted_index.networks,	123
abstract_inverted_index.promising	38
abstract_inverted_index.resulting	175
abstract_inverted_index.training.	27, 85
abstract_inverted_index.typically	4
abstract_inverted_index.end-to-end	7, 218
abstract_inverted_index.facilitate	165
abstract_inverted_index.inadequate	177
abstract_inverted_index.integrated	205
abstract_inverted_index.mitigating	172
abstract_inverted_index.seamlessly	204
abstract_inverted_index.simplistic	134
abstract_inverted_index.surpassing	217
abstract_inverted_index.Exponential	159
abstract_inverted_index.alleviating	124
abstract_inverted_index.challenges,	14
abstract_inverted_index.experiments	190
abstract_inverted_index.frameworks,	210
abstract_inverted_index.independent	119
abstract_inverted_index.information	70, 166
abstract_inverted_index.interaction	68
abstract_inverted_index.performance	80, 125, 214
abstract_inverted_index.Multilaminar	93, 102
abstract_inverted_index.consumption,	19
abstract_inverted_index.consumption.	227
abstract_inverted_index.demonstrated	186
abstract_inverted_index.difficulties	22
abstract_inverted_index.information.	144
abstract_inverted_index.insufficient	129
abstract_inverted_index.interaction.	178
abstract_inverted_index.limitations,	89
abstract_inverted_index.partitioning	46
abstract_inverted_index.performance.	188
abstract_inverted_index.Specifically,	99
abstract_inverted_index.inefficiency,	20
abstract_inverted_index.significantly	211
abstract_inverted_index.parallelization	25
abstract_inverted_index.shortsightedness	174
abstract_inverted_index.gradient-isolated	51
cited_by_percentile_year
countries_distinct_count	0
institutions_distinct_count	8
citation_normalized_percentile.value	0.12515803
citation_normalized_percentile.is_in_top_1_percent	False
citation_normalized_percentile.is_in_top_10_percent	False