LEGO-Learn: Label-Efficient Graph Open-Set Learning Article Swipe
Haoyan Xu
,
Kay Liu
,
Z. P. Yao
,
Philip S. Yu
,
Kaize Ding
,
Yue Zhao
·
YOU?
·
· 2024
· Open Access
·
· DOI: https://doi.org/10.48550/arxiv.2410.16386
YOU?
·
· 2024
· Open Access
·
· DOI: https://doi.org/10.48550/arxiv.2410.16386
How can we train graph-based models to recognize unseen classes while keeping labeling costs low? Graph open-set learning (GOL) and out-of-distribution (OOD) detection aim to address this challenge by training models that can accurately classify known, in-distribution (ID) classes while identifying and handling previously unseen classes during inference. It is critical for high-stakes, real-world applications where models frequently encounter unexpected data, including finance, security, and healthcare. However, current GOL methods assume access to many labeled ID samples, which is unrealistic for large-scale graphs due to high annotation costs. In this paper, we propose LEGO-Learn (Label-Efficient Graph Open-set Learning), a novel framework that tackles open-set node classification on graphs within a given label budget by selecting the most informative ID nodes. LEGO-Learn employs a GNN-based filter to identify and exclude potential OOD nodes and then select highly informative ID nodes for labeling using the K-Medoids algorithm. To prevent the filter from discarding valuable ID examples, we introduce a classifier that differentiates between the C known ID classes and an additional class representing OOD nodes (hence, a C+1 classifier). This classifier uses a weighted cross-entropy loss to balance the removal of OOD nodes while retaining informative ID nodes. Experimental results on four real-world datasets demonstrate that LEGO-Learn significantly outperforms leading methods, with up to a 6.62% improvement in ID classification accuracy and a 7.49% increase in AUROC for OOD detection.
Related Topics
Concepts
Metadata
- Type
- preprint
- Language
- en
- Landing Page
- http://arxiv.org/abs/2410.16386
- https://arxiv.org/pdf/2410.16386
- OA Status
- green
- Related Works
- 10
- OpenAlex ID
- https://openalex.org/W4404260817
All OpenAlex metadata
Raw OpenAlex JSON
- OpenAlex ID
-
https://openalex.org/W4404260817Canonical identifier for this work in OpenAlex
- DOI
-
https://doi.org/10.48550/arxiv.2410.16386Digital Object Identifier
- Title
-
LEGO-Learn: Label-Efficient Graph Open-Set LearningWork title
- Type
-
preprintOpenAlex work type
- Language
-
enPrimary language
- Publication year
-
2024Year of publication
- Publication date
-
2024-10-21Full publication date if available
- Authors
-
Haoyan Xu, Kay Liu, Z. P. Yao, Philip S. Yu, Kaize Ding, Yue ZhaoList of authors in order
- Landing page
-
https://arxiv.org/abs/2410.16386Publisher landing page
- PDF URL
-
https://arxiv.org/pdf/2410.16386Direct link to full text PDF
- Open access
-
YesWhether a free full text is available
- OA status
-
greenOpen access status per OpenAlex
- OA URL
-
https://arxiv.org/pdf/2410.16386Direct OA link when available
- Concepts
-
Computer science, Graph, Artificial intelligence, Set (abstract data type), Machine learning, Theoretical computer science, Programming languageTop concepts (fields/topics) attached by OpenAlex
- Cited by
-
0Total citation count in OpenAlex
- Related works (count)
-
10Other works algorithmically related by OpenAlex
Full payload
| id | https://openalex.org/W4404260817 |
|---|---|
| doi | https://doi.org/10.48550/arxiv.2410.16386 |
| ids.doi | https://doi.org/10.48550/arxiv.2410.16386 |
| ids.openalex | https://openalex.org/W4404260817 |
| fwci | |
| type | preprint |
| title | LEGO-Learn: Label-Efficient Graph Open-Set Learning |
| biblio.issue | |
| biblio.volume | |
| biblio.last_page | |
| biblio.first_page | |
| topics[0].id | https://openalex.org/T11550 |
| topics[0].field.id | https://openalex.org/fields/17 |
| topics[0].field.display_name | Computer Science |
| topics[0].score | 0.9735999703407288 |
| topics[0].domain.id | https://openalex.org/domains/3 |
| topics[0].domain.display_name | Physical Sciences |
| topics[0].subfield.id | https://openalex.org/subfields/1702 |
| topics[0].subfield.display_name | Artificial Intelligence |
| topics[0].display_name | Text and Document Classification Technologies |
| topics[1].id | https://openalex.org/T12535 |
| topics[1].field.id | https://openalex.org/fields/17 |
| topics[1].field.display_name | Computer Science |
| topics[1].score | 0.9344000220298767 |
| 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 | Machine Learning and Data Classification |
| topics[2].id | https://openalex.org/T11273 |
| topics[2].field.id | https://openalex.org/fields/17 |
| topics[2].field.display_name | Computer Science |
| topics[2].score | 0.9171000123023987 |
| topics[2].domain.id | https://openalex.org/domains/3 |
| topics[2].domain.display_name | Physical Sciences |
| topics[2].subfield.id | https://openalex.org/subfields/1702 |
| topics[2].subfield.display_name | Artificial Intelligence |
| topics[2].display_name | Advanced Graph Neural Networks |
| is_xpac | False |
| apc_list | |
| apc_paid | |
| concepts[0].id | https://openalex.org/C41008148 |
| concepts[0].level | 0 |
| concepts[0].score | 0.5979357957839966 |
| concepts[0].wikidata | https://www.wikidata.org/wiki/Q21198 |
| concepts[0].display_name | Computer science |
| concepts[1].id | https://openalex.org/C132525143 |
| concepts[1].level | 2 |
| concepts[1].score | 0.5319203734397888 |
| concepts[1].wikidata | https://www.wikidata.org/wiki/Q141488 |
| concepts[1].display_name | Graph |
| concepts[2].id | https://openalex.org/C154945302 |
| concepts[2].level | 1 |
| concepts[2].score | 0.42758530378341675 |
| concepts[2].wikidata | https://www.wikidata.org/wiki/Q11660 |
| concepts[2].display_name | Artificial intelligence |
| concepts[3].id | https://openalex.org/C177264268 |
| concepts[3].level | 2 |
| concepts[3].score | 0.41871315240859985 |
| concepts[3].wikidata | https://www.wikidata.org/wiki/Q1514741 |
| concepts[3].display_name | Set (abstract data type) |
| concepts[4].id | https://openalex.org/C119857082 |
| concepts[4].level | 1 |
| concepts[4].score | 0.33466845750808716 |
| concepts[4].wikidata | https://www.wikidata.org/wiki/Q2539 |
| concepts[4].display_name | Machine learning |
| concepts[5].id | https://openalex.org/C80444323 |
| concepts[5].level | 1 |
| concepts[5].score | 0.22077032923698425 |
| concepts[5].wikidata | https://www.wikidata.org/wiki/Q2878974 |
| concepts[5].display_name | Theoretical computer science |
| concepts[6].id | https://openalex.org/C199360897 |
| concepts[6].level | 1 |
| concepts[6].score | 0.1044614315032959 |
| concepts[6].wikidata | https://www.wikidata.org/wiki/Q9143 |
| concepts[6].display_name | Programming language |
| keywords[0].id | https://openalex.org/keywords/computer-science |
| keywords[0].score | 0.5979357957839966 |
| keywords[0].display_name | Computer science |
| keywords[1].id | https://openalex.org/keywords/graph |
| keywords[1].score | 0.5319203734397888 |
| keywords[1].display_name | Graph |
| keywords[2].id | https://openalex.org/keywords/artificial-intelligence |
| keywords[2].score | 0.42758530378341675 |
| keywords[2].display_name | Artificial intelligence |
| keywords[3].id | https://openalex.org/keywords/set |
| keywords[3].score | 0.41871315240859985 |
| keywords[3].display_name | Set (abstract data type) |
| keywords[4].id | https://openalex.org/keywords/machine-learning |
| keywords[4].score | 0.33466845750808716 |
| keywords[4].display_name | Machine learning |
| keywords[5].id | https://openalex.org/keywords/theoretical-computer-science |
| keywords[5].score | 0.22077032923698425 |
| keywords[5].display_name | Theoretical computer science |
| keywords[6].id | https://openalex.org/keywords/programming-language |
| keywords[6].score | 0.1044614315032959 |
| keywords[6].display_name | Programming language |
| language | en |
| locations[0].id | pmh:oai:arXiv.org:2410.16386 |
| locations[0].is_oa | True |
| locations[0].source.id | https://openalex.org/S4306400194 |
| locations[0].source.issn | |
| locations[0].source.type | repository |
| locations[0].source.is_oa | True |
| locations[0].source.issn_l | |
| locations[0].source.is_core | False |
| locations[0].source.is_in_doaj | False |
| locations[0].source.display_name | arXiv (Cornell University) |
| locations[0].source.host_organization | https://openalex.org/I205783295 |
| locations[0].source.host_organization_name | Cornell University |
| locations[0].source.host_organization_lineage | https://openalex.org/I205783295 |
| locations[0].license | |
| locations[0].pdf_url | https://arxiv.org/pdf/2410.16386 |
| locations[0].version | submittedVersion |
| locations[0].raw_type | text |
| locations[0].license_id | |
| locations[0].is_accepted | False |
| locations[0].is_published | False |
| locations[0].raw_source_name | |
| locations[0].landing_page_url | http://arxiv.org/abs/2410.16386 |
| locations[1].id | doi:10.48550/arxiv.2410.16386 |
| locations[1].is_oa | True |
| locations[1].source.id | https://openalex.org/S4306400194 |
| locations[1].source.issn | |
| locations[1].source.type | repository |
| locations[1].source.is_oa | True |
| locations[1].source.issn_l | |
| locations[1].source.is_core | False |
| locations[1].source.is_in_doaj | False |
| locations[1].source.display_name | arXiv (Cornell University) |
| locations[1].source.host_organization | https://openalex.org/I205783295 |
| locations[1].source.host_organization_name | Cornell University |
| locations[1].source.host_organization_lineage | https://openalex.org/I205783295 |
| locations[1].license | cc-by |
| locations[1].pdf_url | |
| locations[1].version | |
| locations[1].raw_type | article |
| locations[1].license_id | https://openalex.org/licenses/cc-by |
| locations[1].is_accepted | False |
| locations[1].is_published | |
| locations[1].raw_source_name | |
| locations[1].landing_page_url | https://doi.org/10.48550/arxiv.2410.16386 |
| indexed_in | arxiv, datacite |
| authorships[0].author.id | https://openalex.org/A5110133782 |
| authorships[0].author.orcid | https://orcid.org/0009-0005-3103-8639 |
| authorships[0].author.display_name | Haoyan Xu |
| authorships[0].author_position | first |
| authorships[0].raw_author_name | Xu, Haoyan |
| authorships[0].is_corresponding | False |
| authorships[1].author.id | https://openalex.org/A5070129944 |
| authorships[1].author.orcid | https://orcid.org/0000-0002-2022-9465 |
| authorships[1].author.display_name | Kay Liu |
| authorships[1].author_position | middle |
| authorships[1].raw_author_name | Liu, Kay |
| authorships[1].is_corresponding | False |
| authorships[2].author.id | https://openalex.org/A5069333345 |
| authorships[2].author.orcid | https://orcid.org/0009-0001-7110-3985 |
| authorships[2].author.display_name | Z. P. Yao |
| authorships[2].author_position | middle |
| authorships[2].raw_author_name | Yao, Zhengtao |
| authorships[2].is_corresponding | False |
| authorships[3].author.id | https://openalex.org/A5036357902 |
| authorships[3].author.orcid | https://orcid.org/0000-0002-3491-5968 |
| authorships[3].author.display_name | Philip S. Yu |
| authorships[3].author_position | middle |
| authorships[3].raw_author_name | Yu, Philip S. |
| authorships[3].is_corresponding | False |
| authorships[4].author.id | https://openalex.org/A5044455276 |
| authorships[4].author.orcid | https://orcid.org/0000-0001-6684-6752 |
| authorships[4].author.display_name | Kaize Ding |
| authorships[4].author_position | middle |
| authorships[4].raw_author_name | Ding, Kaize |
| authorships[4].is_corresponding | False |
| authorships[5].author.id | https://openalex.org/A5101429224 |
| authorships[5].author.orcid | https://orcid.org/0000-0003-4022-9208 |
| authorships[5].author.display_name | Yue Zhao |
| authorships[5].author_position | last |
| authorships[5].raw_author_name | Zhao, Yue |
| authorships[5].is_corresponding | False |
| has_content.pdf | False |
| has_content.grobid_xml | False |
| is_paratext | False |
| open_access.is_oa | True |
| open_access.oa_url | https://arxiv.org/pdf/2410.16386 |
| open_access.oa_status | green |
| open_access.any_repository_has_fulltext | False |
| created_date | 2025-10-10T00:00:00 |
| display_name | LEGO-Learn: Label-Efficient Graph Open-Set Learning |
| has_fulltext | False |
| is_retracted | False |
| updated_date | 2025-11-06T06:51:31.235846 |
| primary_topic.id | https://openalex.org/T11550 |
| primary_topic.field.id | https://openalex.org/fields/17 |
| primary_topic.field.display_name | Computer Science |
| primary_topic.score | 0.9735999703407288 |
| primary_topic.domain.id | https://openalex.org/domains/3 |
| primary_topic.domain.display_name | Physical Sciences |
| primary_topic.subfield.id | https://openalex.org/subfields/1702 |
| primary_topic.subfield.display_name | Artificial Intelligence |
| primary_topic.display_name | Text and Document Classification Technologies |
| related_works | https://openalex.org/W2961085424, https://openalex.org/W4306674287, https://openalex.org/W3046775127, https://openalex.org/W4394896187, https://openalex.org/W3170094116, https://openalex.org/W4386462264, https://openalex.org/W3107602296, https://openalex.org/W4364306694, https://openalex.org/W4312192474, https://openalex.org/W4283697347 |
| cited_by_count | 0 |
| locations_count | 2 |
| best_oa_location.id | pmh:oai:arXiv.org:2410.16386 |
| best_oa_location.is_oa | True |
| best_oa_location.source.id | https://openalex.org/S4306400194 |
| best_oa_location.source.issn | |
| best_oa_location.source.type | repository |
| best_oa_location.source.is_oa | True |
| best_oa_location.source.issn_l | |
| best_oa_location.source.is_core | False |
| best_oa_location.source.is_in_doaj | False |
| best_oa_location.source.display_name | arXiv (Cornell University) |
| best_oa_location.source.host_organization | https://openalex.org/I205783295 |
| best_oa_location.source.host_organization_name | Cornell University |
| best_oa_location.source.host_organization_lineage | https://openalex.org/I205783295 |
| best_oa_location.license | |
| best_oa_location.pdf_url | https://arxiv.org/pdf/2410.16386 |
| best_oa_location.version | submittedVersion |
| best_oa_location.raw_type | text |
| best_oa_location.license_id | |
| best_oa_location.is_accepted | False |
| best_oa_location.is_published | False |
| best_oa_location.raw_source_name | |
| best_oa_location.landing_page_url | http://arxiv.org/abs/2410.16386 |
| primary_location.id | pmh:oai:arXiv.org:2410.16386 |
| primary_location.is_oa | True |
| primary_location.source.id | https://openalex.org/S4306400194 |
| primary_location.source.issn | |
| primary_location.source.type | repository |
| primary_location.source.is_oa | True |
| primary_location.source.issn_l | |
| primary_location.source.is_core | False |
| primary_location.source.is_in_doaj | False |
| primary_location.source.display_name | arXiv (Cornell University) |
| primary_location.source.host_organization | https://openalex.org/I205783295 |
| primary_location.source.host_organization_name | Cornell University |
| primary_location.source.host_organization_lineage | https://openalex.org/I205783295 |
| primary_location.license | |
| primary_location.pdf_url | https://arxiv.org/pdf/2410.16386 |
| primary_location.version | submittedVersion |
| primary_location.raw_type | text |
| primary_location.license_id | |
| primary_location.is_accepted | False |
| primary_location.is_published | False |
| primary_location.raw_source_name | |
| primary_location.landing_page_url | http://arxiv.org/abs/2410.16386 |
| publication_date | 2024-10-21 |
| publication_year | 2024 |
| referenced_works_count | 0 |
| abstract_inverted_index.C | 162 |
| abstract_inverted_index.a | 98, 109, 122, 156, 174, 180, 212, 220 |
| abstract_inverted_index.ID | 75, 118, 137, 152, 164, 194, 216 |
| abstract_inverted_index.In | 88 |
| abstract_inverted_index.It | 48 |
| abstract_inverted_index.To | 145 |
| abstract_inverted_index.an | 167 |
| abstract_inverted_index.by | 28, 113 |
| abstract_inverted_index.in | 215, 223 |
| abstract_inverted_index.is | 49, 78 |
| abstract_inverted_index.of | 188 |
| abstract_inverted_index.on | 106, 198 |
| abstract_inverted_index.to | 6, 24, 72, 84, 125, 184, 211 |
| abstract_inverted_index.up | 210 |
| abstract_inverted_index.we | 2, 91, 154 |
| abstract_inverted_index.C+1 | 175 |
| abstract_inverted_index.GOL | 68 |
| abstract_inverted_index.How | 0 |
| abstract_inverted_index.OOD | 130, 171, 189, 226 |
| abstract_inverted_index.aim | 23 |
| abstract_inverted_index.and | 19, 41, 64, 127, 132, 166, 219 |
| abstract_inverted_index.can | 1, 32 |
| abstract_inverted_index.due | 83 |
| abstract_inverted_index.for | 51, 80, 139, 225 |
| abstract_inverted_index.the | 115, 142, 147, 161, 186 |
| abstract_inverted_index.(ID) | 37 |
| abstract_inverted_index.This | 177 |
| abstract_inverted_index.four | 199 |
| abstract_inverted_index.from | 149 |
| abstract_inverted_index.high | 85 |
| abstract_inverted_index.loss | 183 |
| abstract_inverted_index.low? | 14 |
| abstract_inverted_index.many | 73 |
| abstract_inverted_index.most | 116 |
| abstract_inverted_index.node | 104 |
| abstract_inverted_index.that | 31, 101, 158, 203 |
| abstract_inverted_index.then | 133 |
| abstract_inverted_index.this | 26, 89 |
| abstract_inverted_index.uses | 179 |
| abstract_inverted_index.with | 209 |
| abstract_inverted_index.(GOL) | 18 |
| abstract_inverted_index.(OOD) | 21 |
| abstract_inverted_index.6.62% | 213 |
| abstract_inverted_index.7.49% | 221 |
| abstract_inverted_index.AUROC | 224 |
| abstract_inverted_index.Graph | 15, 95 |
| abstract_inverted_index.class | 169 |
| abstract_inverted_index.costs | 13 |
| abstract_inverted_index.data, | 60 |
| abstract_inverted_index.given | 110 |
| abstract_inverted_index.known | 163 |
| abstract_inverted_index.label | 111 |
| abstract_inverted_index.nodes | 131, 138, 172, 190 |
| abstract_inverted_index.novel | 99 |
| abstract_inverted_index.train | 3 |
| abstract_inverted_index.using | 141 |
| abstract_inverted_index.where | 55 |
| abstract_inverted_index.which | 77 |
| abstract_inverted_index.while | 10, 39, 191 |
| abstract_inverted_index.access | 71 |
| abstract_inverted_index.assume | 70 |
| abstract_inverted_index.budget | 112 |
| abstract_inverted_index.costs. | 87 |
| abstract_inverted_index.during | 46 |
| abstract_inverted_index.filter | 124, 148 |
| abstract_inverted_index.graphs | 82, 107 |
| abstract_inverted_index.highly | 135 |
| abstract_inverted_index.known, | 35 |
| abstract_inverted_index.models | 5, 30, 56 |
| abstract_inverted_index.nodes. | 119, 195 |
| abstract_inverted_index.paper, | 90 |
| abstract_inverted_index.select | 134 |
| abstract_inverted_index.unseen | 8, 44 |
| abstract_inverted_index.within | 108 |
| abstract_inverted_index.(hence, | 173 |
| abstract_inverted_index.address | 25 |
| abstract_inverted_index.balance | 185 |
| abstract_inverted_index.between | 160 |
| abstract_inverted_index.classes | 9, 38, 45, 165 |
| abstract_inverted_index.current | 67 |
| abstract_inverted_index.employs | 121 |
| abstract_inverted_index.exclude | 128 |
| abstract_inverted_index.keeping | 11 |
| abstract_inverted_index.labeled | 74 |
| abstract_inverted_index.leading | 207 |
| abstract_inverted_index.methods | 69 |
| abstract_inverted_index.prevent | 146 |
| abstract_inverted_index.propose | 92 |
| abstract_inverted_index.removal | 187 |
| abstract_inverted_index.results | 197 |
| abstract_inverted_index.tackles | 102 |
| abstract_inverted_index.However, | 66 |
| abstract_inverted_index.Open-set | 96 |
| abstract_inverted_index.accuracy | 218 |
| abstract_inverted_index.classify | 34 |
| abstract_inverted_index.critical | 50 |
| abstract_inverted_index.datasets | 201 |
| abstract_inverted_index.finance, | 62 |
| abstract_inverted_index.handling | 42 |
| abstract_inverted_index.identify | 126 |
| abstract_inverted_index.increase | 222 |
| abstract_inverted_index.labeling | 12, 140 |
| abstract_inverted_index.learning | 17 |
| abstract_inverted_index.methods, | 208 |
| abstract_inverted_index.open-set | 16, 103 |
| abstract_inverted_index.samples, | 76 |
| abstract_inverted_index.training | 29 |
| abstract_inverted_index.valuable | 151 |
| abstract_inverted_index.weighted | 181 |
| abstract_inverted_index.GNN-based | 123 |
| abstract_inverted_index.K-Medoids | 143 |
| abstract_inverted_index.challenge | 27 |
| abstract_inverted_index.detection | 22 |
| abstract_inverted_index.encounter | 58 |
| abstract_inverted_index.examples, | 153 |
| abstract_inverted_index.framework | 100 |
| abstract_inverted_index.including | 61 |
| abstract_inverted_index.introduce | 155 |
| abstract_inverted_index.potential | 129 |
| abstract_inverted_index.recognize | 7 |
| abstract_inverted_index.retaining | 192 |
| abstract_inverted_index.security, | 63 |
| abstract_inverted_index.selecting | 114 |
| abstract_inverted_index.LEGO-Learn | 93, 120, 204 |
| abstract_inverted_index.Learning), | 97 |
| abstract_inverted_index.accurately | 33 |
| abstract_inverted_index.additional | 168 |
| abstract_inverted_index.algorithm. | 144 |
| abstract_inverted_index.annotation | 86 |
| abstract_inverted_index.classifier | 157, 178 |
| abstract_inverted_index.detection. | 227 |
| abstract_inverted_index.discarding | 150 |
| abstract_inverted_index.frequently | 57 |
| abstract_inverted_index.inference. | 47 |
| abstract_inverted_index.previously | 43 |
| abstract_inverted_index.real-world | 53, 200 |
| abstract_inverted_index.unexpected | 59 |
| abstract_inverted_index.demonstrate | 202 |
| abstract_inverted_index.graph-based | 4 |
| abstract_inverted_index.healthcare. | 65 |
| abstract_inverted_index.identifying | 40 |
| abstract_inverted_index.improvement | 214 |
| abstract_inverted_index.informative | 117, 136, 193 |
| abstract_inverted_index.large-scale | 81 |
| abstract_inverted_index.outperforms | 206 |
| abstract_inverted_index.unrealistic | 79 |
| abstract_inverted_index.Experimental | 196 |
| abstract_inverted_index.applications | 54 |
| abstract_inverted_index.classifier). | 176 |
| abstract_inverted_index.high-stakes, | 52 |
| abstract_inverted_index.representing | 170 |
| abstract_inverted_index.cross-entropy | 182 |
| abstract_inverted_index.significantly | 205 |
| abstract_inverted_index.classification | 105, 217 |
| abstract_inverted_index.differentiates | 159 |
| abstract_inverted_index.in-distribution | 36 |
| abstract_inverted_index.(Label-Efficient | 94 |
| abstract_inverted_index.out-of-distribution | 20 |
| cited_by_percentile_year | |
| countries_distinct_count | 0 |
| institutions_distinct_count | 6 |
| citation_normalized_percentile |