Robust partial-label learning by leveraging class activation values Article Swipe
Tobias A. Fuchs
,
Florian Kalinke
·
YOU?
·
· 2025
· Open Access
·
· DOI: https://doi.org/10.1007/s10994-025-06796-z
YOU?
·
· 2025
· Open Access
·
· DOI: https://doi.org/10.1007/s10994-025-06796-z
Real-world training data is often noisy; for example, human annotators assign conflicting class labels to the same instances. Partial-label learning (PLL) is a weakly supervised learning paradigm that allows training classifiers in this context without manual data cleaning. While state-of-the-art methods have good predictive performance, their predictions are sensitive to high noise levels, out-of-distribution data, and adversarial perturbations. We propose a novel PLL method based on subjective logic, which explicitly represents uncertainty by leveraging the magnitudes of the underlying neural network’s class activation values. Thereby, we effectively incorporate prior knowledge about the class labels by using a novel label weight re-distribution strategy that we prove to be optimal. We empirically show that our method yields more robust predictions in terms of predictive performance under high PLL noise levels, handling out-of-distribution examples, and handling adversarial perturbations on the test instances.
Related Topics
Metadata
- Type
- article
- Language
- en
- Landing Page
- https://doi.org/10.1007/s10994-025-06796-z
- https://link.springer.com/content/pdf/10.1007/s10994-025-06796-z.pdf
- OA Status
- hybrid
- References
- 59
- Related Works
- 10
- OpenAlex ID
- https://openalex.org/W4412441224
All OpenAlex metadata
Raw OpenAlex JSON
- OpenAlex ID
-
https://openalex.org/W4412441224Canonical identifier for this work in OpenAlex
- DOI
-
https://doi.org/10.1007/s10994-025-06796-zDigital Object Identifier
- Title
-
Robust partial-label learning by leveraging class activation valuesWork title
- Type
-
articleOpenAlex work type
- Language
-
enPrimary language
- Publication year
-
2025Year of publication
- Publication date
-
2025-07-15Full publication date if available
- Authors
-
Tobias A. Fuchs, Florian KalinkeList of authors in order
- Landing page
-
https://doi.org/10.1007/s10994-025-06796-zPublisher landing page
- PDF URL
-
https://link.springer.com/content/pdf/10.1007/s10994-025-06796-z.pdfDirect link to full text PDF
- Open access
-
YesWhether a free full text is available
- OA status
-
hybridOpen access status per OpenAlex
- OA URL
-
https://link.springer.com/content/pdf/10.1007/s10994-025-06796-z.pdfDirect OA link when available
- Concepts
-
Class (philosophy), Artificial intelligence, Computer science, Machine learning, MathematicsTop concepts (fields/topics) attached by OpenAlex
- Cited by
-
0Total citation count in OpenAlex
- References (count)
-
59Number of works referenced by this work
- Related works (count)
-
10Other works algorithmically related by OpenAlex
Full payload
| id | https://openalex.org/W4412441224 |
|---|---|
| doi | https://doi.org/10.1007/s10994-025-06796-z |
| ids.doi | https://doi.org/10.1007/s10994-025-06796-z |
| ids.openalex | https://openalex.org/W4412441224 |
| fwci | 0.0 |
| type | article |
| title | Robust partial-label learning by leveraging class activation values |
| biblio.issue | 8 |
| biblio.volume | 114 |
| 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 | 0.9950000047683716 |
| 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/T10601 |
| topics[1].field.id | https://openalex.org/fields/17 |
| topics[1].field.display_name | Computer Science |
| topics[1].score | 0.9868999719619751 |
| topics[1].domain.id | https://openalex.org/domains/3 |
| topics[1].domain.display_name | Physical Sciences |
| topics[1].subfield.id | https://openalex.org/subfields/1707 |
| topics[1].subfield.display_name | Computer Vision and Pattern Recognition |
| topics[1].display_name | Handwritten Text Recognition Techniques |
| topics[2].id | https://openalex.org/T12254 |
| topics[2].field.id | https://openalex.org/fields/13 |
| topics[2].field.display_name | Biochemistry, Genetics and Molecular Biology |
| topics[2].score | 0.9509000182151794 |
| topics[2].domain.id | https://openalex.org/domains/1 |
| topics[2].domain.display_name | Life Sciences |
| topics[2].subfield.id | https://openalex.org/subfields/1312 |
| topics[2].subfield.display_name | Molecular Biology |
| topics[2].display_name | Machine Learning in Bioinformatics |
| is_xpac | False |
| apc_list.value | 2390 |
| apc_list.currency | EUR |
| apc_list.value_usd | 2990 |
| apc_paid.value | 2390 |
| apc_paid.currency | EUR |
| apc_paid.value_usd | 2990 |
| concepts[0].id | https://openalex.org/C2777212361 |
| concepts[0].level | 2 |
| concepts[0].score | 0.6174358129501343 |
| concepts[0].wikidata | https://www.wikidata.org/wiki/Q5127848 |
| concepts[0].display_name | Class (philosophy) |
| concepts[1].id | https://openalex.org/C154945302 |
| concepts[1].level | 1 |
| concepts[1].score | 0.5284115076065063 |
| concepts[1].wikidata | https://www.wikidata.org/wiki/Q11660 |
| concepts[1].display_name | Artificial intelligence |
| concepts[2].id | https://openalex.org/C41008148 |
| concepts[2].level | 0 |
| concepts[2].score | 0.5108572840690613 |
| concepts[2].wikidata | https://www.wikidata.org/wiki/Q21198 |
| concepts[2].display_name | Computer science |
| concepts[3].id | https://openalex.org/C119857082 |
| concepts[3].level | 1 |
| concepts[3].score | 0.4373813271522522 |
| concepts[3].wikidata | https://www.wikidata.org/wiki/Q2539 |
| concepts[3].display_name | Machine learning |
| concepts[4].id | https://openalex.org/C33923547 |
| concepts[4].level | 0 |
| concepts[4].score | 0.33952558040618896 |
| concepts[4].wikidata | https://www.wikidata.org/wiki/Q395 |
| concepts[4].display_name | Mathematics |
| keywords[0].id | https://openalex.org/keywords/class |
| keywords[0].score | 0.6174358129501343 |
| keywords[0].display_name | Class (philosophy) |
| keywords[1].id | https://openalex.org/keywords/artificial-intelligence |
| keywords[1].score | 0.5284115076065063 |
| keywords[1].display_name | Artificial intelligence |
| keywords[2].id | https://openalex.org/keywords/computer-science |
| keywords[2].score | 0.5108572840690613 |
| keywords[2].display_name | Computer science |
| keywords[3].id | https://openalex.org/keywords/machine-learning |
| keywords[3].score | 0.4373813271522522 |
| keywords[3].display_name | Machine learning |
| keywords[4].id | https://openalex.org/keywords/mathematics |
| keywords[4].score | 0.33952558040618896 |
| keywords[4].display_name | Mathematics |
| language | en |
| locations[0].id | doi:10.1007/s10994-025-06796-z |
| locations[0].is_oa | True |
| locations[0].source.id | https://openalex.org/S62148650 |
| locations[0].source.issn | 0885-6125, 1573-0565 |
| locations[0].source.type | journal |
| locations[0].source.is_oa | False |
| locations[0].source.issn_l | 0885-6125 |
| locations[0].source.is_core | True |
| locations[0].source.is_in_doaj | False |
| locations[0].source.display_name | Machine Learning |
| locations[0].source.host_organization | https://openalex.org/P4310319900 |
| locations[0].source.host_organization_name | Springer Science+Business Media |
| locations[0].source.host_organization_lineage | https://openalex.org/P4310319900, https://openalex.org/P4310319965 |
| locations[0].source.host_organization_lineage_names | Springer Science+Business Media, Springer Nature |
| locations[0].license | cc-by |
| locations[0].pdf_url | https://link.springer.com/content/pdf/10.1007/s10994-025-06796-z.pdf |
| locations[0].version | publishedVersion |
| locations[0].raw_type | journal-article |
| locations[0].license_id | https://openalex.org/licenses/cc-by |
| locations[0].is_accepted | True |
| locations[0].is_published | True |
| locations[0].raw_source_name | Machine Learning |
| locations[0].landing_page_url | https://doi.org/10.1007/s10994-025-06796-z |
| indexed_in | crossref |
| authorships[0].author.id | https://openalex.org/A5013911322 |
| authorships[0].author.orcid | https://orcid.org/0000-0001-8812-9359 |
| authorships[0].author.display_name | Tobias A. Fuchs |
| authorships[0].countries | DE |
| authorships[0].affiliations[0].institution_ids | https://openalex.org/I102335020 |
| authorships[0].affiliations[0].raw_affiliation_string | Karlsruhe Institute of Technology, Karlsruhe, Germany |
| authorships[0].institutions[0].id | https://openalex.org/I102335020 |
| authorships[0].institutions[0].ror | https://ror.org/04t3en479 |
| authorships[0].institutions[0].type | education |
| authorships[0].institutions[0].lineage | https://openalex.org/I102335020, https://openalex.org/I1305996414 |
| authorships[0].institutions[0].country_code | DE |
| authorships[0].institutions[0].display_name | Karlsruhe Institute of Technology |
| authorships[0].author_position | first |
| authorships[0].raw_author_name | Tobias Fuchs |
| authorships[0].is_corresponding | False |
| authorships[0].raw_affiliation_strings | Karlsruhe Institute of Technology, Karlsruhe, Germany |
| authorships[1].author.id | https://openalex.org/A5054748088 |
| authorships[1].author.orcid | https://orcid.org/0000-0002-0443-6288 |
| authorships[1].author.display_name | Florian Kalinke |
| authorships[1].countries | DE |
| authorships[1].affiliations[0].institution_ids | https://openalex.org/I102335020 |
| authorships[1].affiliations[0].raw_affiliation_string | Karlsruhe Institute of Technology, Karlsruhe, Germany |
| authorships[1].institutions[0].id | https://openalex.org/I102335020 |
| authorships[1].institutions[0].ror | https://ror.org/04t3en479 |
| authorships[1].institutions[0].type | education |
| authorships[1].institutions[0].lineage | https://openalex.org/I102335020, https://openalex.org/I1305996414 |
| authorships[1].institutions[0].country_code | DE |
| authorships[1].institutions[0].display_name | Karlsruhe Institute of Technology |
| authorships[1].author_position | last |
| authorships[1].raw_author_name | Florian Kalinke |
| authorships[1].is_corresponding | False |
| authorships[1].raw_affiliation_strings | Karlsruhe Institute of Technology, Karlsruhe, Germany |
| has_content.pdf | True |
| has_content.grobid_xml | True |
| is_paratext | False |
| open_access.is_oa | True |
| open_access.oa_url | https://link.springer.com/content/pdf/10.1007/s10994-025-06796-z.pdf |
| open_access.oa_status | hybrid |
| open_access.any_repository_has_fulltext | False |
| created_date | 2025-10-10T00:00:00 |
| display_name | Robust partial-label learning by leveraging class activation values |
| has_fulltext | False |
| is_retracted | False |
| updated_date | 2025-11-06T03:46:38.306776 |
| 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 | 0.9950000047683716 |
| 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/W4391375266, https://openalex.org/W2961085424, https://openalex.org/W4306674287, https://openalex.org/W4387369504, https://openalex.org/W4394896187, https://openalex.org/W3170094116, https://openalex.org/W4386462264, https://openalex.org/W3107602296, https://openalex.org/W4364306694, https://openalex.org/W4312192474 |
| cited_by_count | 0 |
| locations_count | 1 |
| best_oa_location.id | doi:10.1007/s10994-025-06796-z |
| best_oa_location.is_oa | True |
| best_oa_location.source.id | https://openalex.org/S62148650 |
| best_oa_location.source.issn | 0885-6125, 1573-0565 |
| best_oa_location.source.type | journal |
| best_oa_location.source.is_oa | False |
| best_oa_location.source.issn_l | 0885-6125 |
| best_oa_location.source.is_core | True |
| best_oa_location.source.is_in_doaj | False |
| best_oa_location.source.display_name | Machine Learning |
| best_oa_location.source.host_organization | https://openalex.org/P4310319900 |
| best_oa_location.source.host_organization_name | Springer Science+Business Media |
| best_oa_location.source.host_organization_lineage | https://openalex.org/P4310319900, https://openalex.org/P4310319965 |
| best_oa_location.source.host_organization_lineage_names | Springer Science+Business Media, Springer Nature |
| best_oa_location.license | cc-by |
| best_oa_location.pdf_url | https://link.springer.com/content/pdf/10.1007/s10994-025-06796-z.pdf |
| best_oa_location.version | publishedVersion |
| best_oa_location.raw_type | journal-article |
| best_oa_location.license_id | https://openalex.org/licenses/cc-by |
| best_oa_location.is_accepted | True |
| best_oa_location.is_published | True |
| best_oa_location.raw_source_name | Machine Learning |
| best_oa_location.landing_page_url | https://doi.org/10.1007/s10994-025-06796-z |
| primary_location.id | doi:10.1007/s10994-025-06796-z |
| primary_location.is_oa | True |
| primary_location.source.id | https://openalex.org/S62148650 |
| primary_location.source.issn | 0885-6125, 1573-0565 |
| primary_location.source.type | journal |
| primary_location.source.is_oa | False |
| primary_location.source.issn_l | 0885-6125 |
| primary_location.source.is_core | True |
| primary_location.source.is_in_doaj | False |
| primary_location.source.display_name | Machine Learning |
| primary_location.source.host_organization | https://openalex.org/P4310319900 |
| primary_location.source.host_organization_name | Springer Science+Business Media |
| primary_location.source.host_organization_lineage | https://openalex.org/P4310319900, https://openalex.org/P4310319965 |
| primary_location.source.host_organization_lineage_names | Springer Science+Business Media, Springer Nature |
| primary_location.license | cc-by |
| primary_location.pdf_url | https://link.springer.com/content/pdf/10.1007/s10994-025-06796-z.pdf |
| primary_location.version | publishedVersion |
| primary_location.raw_type | journal-article |
| primary_location.license_id | https://openalex.org/licenses/cc-by |
| primary_location.is_accepted | True |
| primary_location.is_published | True |
| primary_location.raw_source_name | Machine Learning |
| primary_location.landing_page_url | https://doi.org/10.1007/s10994-025-06796-z |
| publication_date | 2025-07-15 |
| publication_year | 2025 |
| referenced_works | https://openalex.org/W4385681639, https://openalex.org/W4310491441, https://openalex.org/W6629510986, https://openalex.org/W2137917285, https://openalex.org/W3008699320, https://openalex.org/W3033991488, https://openalex.org/W2902986194, https://openalex.org/W6683235360, https://openalex.org/W2406470551, https://openalex.org/W4390706868, https://openalex.org/W2903641274, https://openalex.org/W6780619371, https://openalex.org/W4407425788, https://openalex.org/W6913061175, https://openalex.org/W1945616565, https://openalex.org/W2142863987, https://openalex.org/W2194775991, https://openalex.org/W4234173777, https://openalex.org/W2155803963, https://openalex.org/W2146111596, https://openalex.org/W3014596384, https://openalex.org/W6681102160, https://openalex.org/W2811513716, https://openalex.org/W6640963894, https://openalex.org/W2560321925, https://openalex.org/W2130211925, https://openalex.org/W2112796928, https://openalex.org/W2132509897, https://openalex.org/W3007945235, https://openalex.org/W2640329709, https://openalex.org/W2243397390, https://openalex.org/W6839093018, https://openalex.org/W2024328138, https://openalex.org/W3169316096, https://openalex.org/W2790660115, https://openalex.org/W1498436455, https://openalex.org/W2806471870, https://openalex.org/W6662489520, https://openalex.org/W2142623206, https://openalex.org/W2095705004, https://openalex.org/W6637162671, https://openalex.org/W2604553154, https://openalex.org/W4402704591, https://openalex.org/W1573503290, https://openalex.org/W2808448059, https://openalex.org/W4382469817, https://openalex.org/W2750384547, https://openalex.org/W4281626294, https://openalex.org/W2905443329, https://openalex.org/W6803599732, https://openalex.org/W2056671638, https://openalex.org/W2338068721, https://openalex.org/W2106008047, https://openalex.org/W2285223868, https://openalex.org/W3132170818, https://openalex.org/W2733555913, https://openalex.org/W2493420751, https://openalex.org/W4212863985, https://openalex.org/W3134774296 |
| referenced_works_count | 59 |
| abstract_inverted_index.a | 23, 61, 97 |
| abstract_inverted_index.We | 59, 109 |
| abstract_inverted_index.be | 107 |
| abstract_inverted_index.by | 73, 95 |
| abstract_inverted_index.in | 32, 119 |
| abstract_inverted_index.is | 4, 22 |
| abstract_inverted_index.of | 77, 121 |
| abstract_inverted_index.on | 66, 136 |
| abstract_inverted_index.to | 15, 50, 106 |
| abstract_inverted_index.we | 86, 104 |
| abstract_inverted_index.PLL | 63, 126 |
| abstract_inverted_index.and | 56, 132 |
| abstract_inverted_index.are | 48 |
| abstract_inverted_index.for | 7 |
| abstract_inverted_index.our | 113 |
| abstract_inverted_index.the | 16, 75, 78, 92, 137 |
| abstract_inverted_index.data | 3, 37 |
| abstract_inverted_index.good | 43 |
| abstract_inverted_index.have | 42 |
| abstract_inverted_index.high | 51, 125 |
| abstract_inverted_index.more | 116 |
| abstract_inverted_index.same | 17 |
| abstract_inverted_index.show | 111 |
| abstract_inverted_index.test | 138 |
| abstract_inverted_index.that | 28, 103, 112 |
| abstract_inverted_index.this | 33 |
| abstract_inverted_index.(PLL) | 21 |
| abstract_inverted_index.While | 39 |
| abstract_inverted_index.about | 91 |
| abstract_inverted_index.based | 65 |
| abstract_inverted_index.class | 13, 82, 93 |
| abstract_inverted_index.data, | 55 |
| abstract_inverted_index.human | 9 |
| abstract_inverted_index.label | 99 |
| abstract_inverted_index.noise | 52, 127 |
| abstract_inverted_index.novel | 62, 98 |
| abstract_inverted_index.often | 5 |
| abstract_inverted_index.prior | 89 |
| abstract_inverted_index.prove | 105 |
| abstract_inverted_index.terms | 120 |
| abstract_inverted_index.their | 46 |
| abstract_inverted_index.under | 124 |
| abstract_inverted_index.using | 96 |
| abstract_inverted_index.which | 69 |
| abstract_inverted_index.allows | 29 |
| abstract_inverted_index.assign | 11 |
| abstract_inverted_index.labels | 14, 94 |
| abstract_inverted_index.logic, | 68 |
| abstract_inverted_index.manual | 36 |
| abstract_inverted_index.method | 64, 114 |
| abstract_inverted_index.neural | 80 |
| abstract_inverted_index.noisy; | 6 |
| abstract_inverted_index.robust | 117 |
| abstract_inverted_index.weakly | 24 |
| abstract_inverted_index.weight | 100 |
| abstract_inverted_index.yields | 115 |
| abstract_inverted_index.context | 34 |
| abstract_inverted_index.levels, | 53, 128 |
| abstract_inverted_index.methods | 41 |
| abstract_inverted_index.propose | 60 |
| abstract_inverted_index.values. | 84 |
| abstract_inverted_index.without | 35 |
| abstract_inverted_index.Abstract | 0 |
| abstract_inverted_index.Thereby, | 85 |
| abstract_inverted_index.example, | 8 |
| abstract_inverted_index.handling | 129, 133 |
| abstract_inverted_index.learning | 20, 26 |
| abstract_inverted_index.optimal. | 108 |
| abstract_inverted_index.paradigm | 27 |
| abstract_inverted_index.strategy | 102 |
| abstract_inverted_index.training | 2, 30 |
| abstract_inverted_index.cleaning. | 38 |
| abstract_inverted_index.examples, | 131 |
| abstract_inverted_index.knowledge | 90 |
| abstract_inverted_index.sensitive | 49 |
| abstract_inverted_index.Real-world | 1 |
| abstract_inverted_index.activation | 83 |
| abstract_inverted_index.annotators | 10 |
| abstract_inverted_index.explicitly | 70 |
| abstract_inverted_index.instances. | 18, 139 |
| abstract_inverted_index.leveraging | 74 |
| abstract_inverted_index.magnitudes | 76 |
| abstract_inverted_index.predictive | 44, 122 |
| abstract_inverted_index.represents | 71 |
| abstract_inverted_index.subjective | 67 |
| abstract_inverted_index.supervised | 25 |
| abstract_inverted_index.underlying | 79 |
| abstract_inverted_index.adversarial | 57, 134 |
| abstract_inverted_index.classifiers | 31 |
| abstract_inverted_index.conflicting | 12 |
| abstract_inverted_index.effectively | 87 |
| abstract_inverted_index.empirically | 110 |
| abstract_inverted_index.incorporate | 88 |
| abstract_inverted_index.network’s | 81 |
| abstract_inverted_index.performance | 123 |
| abstract_inverted_index.predictions | 47, 118 |
| abstract_inverted_index.uncertainty | 72 |
| abstract_inverted_index.performance, | 45 |
| abstract_inverted_index.Partial-label | 19 |
| abstract_inverted_index.perturbations | 135 |
| abstract_inverted_index.perturbations. | 58 |
| abstract_inverted_index.re-distribution | 101 |
| abstract_inverted_index.state-of-the-art | 40 |
| abstract_inverted_index.out-of-distribution | 54, 130 |
| cited_by_percentile_year | |
| countries_distinct_count | 1 |
| institutions_distinct_count | 2 |
| citation_normalized_percentile.value | 0.12691868 |
| citation_normalized_percentile.is_in_top_1_percent | False |
| citation_normalized_percentile.is_in_top_10_percent | True |