Enhancing Self-Training Methods Article Swipe
Aswathnarayan Radhakrishnan
,
Jim Davis
,
Zachary Rabin
,
Benjamin M. Lewis
,
Matthew Scherreik
,
Roman Ilin
·
YOU?
·
· 2023
· Open Access
·
· DOI: https://doi.org/10.48550/arxiv.2301.07294
YOU?
·
· 2023
· Open Access
·
· DOI: https://doi.org/10.48550/arxiv.2301.07294
Semi-supervised learning approaches train on small sets of labeled data along with large sets of unlabeled data. Self-training is a semi-supervised teacher-student approach that often suffers from the problem of "confirmation bias" that occurs when the student model repeatedly overfits to incorrect pseudo-labels given by the teacher model for the unlabeled data. This bias impedes improvements in pseudo-label accuracy across self-training iterations, leading to unwanted saturation in model performance after just a few iterations. In this work, we describe multiple enhancements to improve the self-training pipeline to mitigate the effect of confirmation bias. We evaluate our enhancements over multiple datasets showing performance gains over existing self-training design choices. Finally, we also study the extendability of our enhanced approach to Open Set unlabeled data (containing classes not seen in labeled data).
Related Topics
Concepts
Metadata
- Type
- preprint
- Language
- en
- Landing Page
- http://arxiv.org/abs/2301.07294
- https://arxiv.org/pdf/2301.07294
- OA Status
- green
- Related Works
- 10
- OpenAlex ID
- https://openalex.org/W4317548404
All OpenAlex metadata
Raw OpenAlex JSON
- OpenAlex ID
-
https://openalex.org/W4317548404Canonical identifier for this work in OpenAlex
- DOI
-
https://doi.org/10.48550/arxiv.2301.07294Digital Object Identifier
- Title
-
Enhancing Self-Training MethodsWork title
- Type
-
preprintOpenAlex work type
- Language
-
enPrimary language
- Publication year
-
2023Year of publication
- Publication date
-
2023-01-18Full publication date if available
- Authors
-
Aswathnarayan Radhakrishnan, Jim Davis, Zachary Rabin, Benjamin M. Lewis, Matthew Scherreik, Roman IlinList of authors in order
- Landing page
-
https://arxiv.org/abs/2301.07294Publisher landing page
- PDF URL
-
https://arxiv.org/pdf/2301.07294Direct 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/2301.07294Direct OA link when available
- Concepts
-
Training set, Labeled data, Computer science, Pipeline (software), Artificial intelligence, Training (meteorology), Machine learning, Set (abstract data type), Data set, Self consistent, Programming language, Quantum electrodynamics, Meteorology, PhysicsTop 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/W4317548404 |
|---|---|
| doi | https://doi.org/10.48550/arxiv.2301.07294 |
| ids.doi | https://doi.org/10.48550/arxiv.2301.07294 |
| ids.openalex | https://openalex.org/W4317548404 |
| fwci | |
| type | preprint |
| title | Enhancing Self-Training Methods |
| biblio.issue | |
| biblio.volume | |
| biblio.last_page | |
| biblio.first_page | |
| topics[0].id | https://openalex.org/T12535 |
| topics[0].field.id | https://openalex.org/fields/17 |
| topics[0].field.display_name | Computer Science |
| topics[0].score | 0.9986000061035156 |
| 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 | Machine Learning and Data Classification |
| topics[1].id | https://openalex.org/T11307 |
| topics[1].field.id | https://openalex.org/fields/17 |
| topics[1].field.display_name | Computer Science |
| topics[1].score | 0.979200005531311 |
| 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 | Domain Adaptation and Few-Shot Learning |
| topics[2].id | https://openalex.org/T11652 |
| topics[2].field.id | https://openalex.org/fields/17 |
| topics[2].field.display_name | Computer Science |
| topics[2].score | 0.9739000201225281 |
| 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 | Imbalanced Data Classification Techniques |
| is_xpac | False |
| apc_list | |
| apc_paid | |
| concepts[0].id | https://openalex.org/C51632099 |
| concepts[0].level | 2 |
| concepts[0].score | 0.7339940071105957 |
| concepts[0].wikidata | https://www.wikidata.org/wiki/Q3985153 |
| concepts[0].display_name | Training set |
| concepts[1].id | https://openalex.org/C2776145971 |
| concepts[1].level | 2 |
| concepts[1].score | 0.7283814549446106 |
| concepts[1].wikidata | https://www.wikidata.org/wiki/Q30673951 |
| concepts[1].display_name | Labeled data |
| concepts[2].id | https://openalex.org/C41008148 |
| concepts[2].level | 0 |
| concepts[2].score | 0.723684549331665 |
| concepts[2].wikidata | https://www.wikidata.org/wiki/Q21198 |
| concepts[2].display_name | Computer science |
| concepts[3].id | https://openalex.org/C43521106 |
| concepts[3].level | 2 |
| concepts[3].score | 0.6110484004020691 |
| concepts[3].wikidata | https://www.wikidata.org/wiki/Q2165493 |
| concepts[3].display_name | Pipeline (software) |
| concepts[4].id | https://openalex.org/C154945302 |
| concepts[4].level | 1 |
| concepts[4].score | 0.594964325428009 |
| concepts[4].wikidata | https://www.wikidata.org/wiki/Q11660 |
| concepts[4].display_name | Artificial intelligence |
| concepts[5].id | https://openalex.org/C2777211547 |
| concepts[5].level | 2 |
| concepts[5].score | 0.5541278719902039 |
| concepts[5].wikidata | https://www.wikidata.org/wiki/Q17141490 |
| concepts[5].display_name | Training (meteorology) |
| concepts[6].id | https://openalex.org/C119857082 |
| concepts[6].level | 1 |
| concepts[6].score | 0.5389244556427002 |
| concepts[6].wikidata | https://www.wikidata.org/wiki/Q2539 |
| concepts[6].display_name | Machine learning |
| concepts[7].id | https://openalex.org/C177264268 |
| concepts[7].level | 2 |
| concepts[7].score | 0.5050604939460754 |
| concepts[7].wikidata | https://www.wikidata.org/wiki/Q1514741 |
| concepts[7].display_name | Set (abstract data type) |
| concepts[8].id | https://openalex.org/C58489278 |
| concepts[8].level | 2 |
| concepts[8].score | 0.4662512540817261 |
| concepts[8].wikidata | https://www.wikidata.org/wiki/Q1172284 |
| concepts[8].display_name | Data set |
| concepts[9].id | https://openalex.org/C2984577571 |
| concepts[9].level | 2 |
| concepts[9].score | 0.41620492935180664 |
| concepts[9].wikidata | https://www.wikidata.org/wiki/Q1319773 |
| concepts[9].display_name | Self consistent |
| concepts[10].id | https://openalex.org/C199360897 |
| concepts[10].level | 1 |
| concepts[10].score | 0.0 |
| concepts[10].wikidata | https://www.wikidata.org/wiki/Q9143 |
| concepts[10].display_name | Programming language |
| concepts[11].id | https://openalex.org/C3079626 |
| concepts[11].level | 1 |
| concepts[11].score | 0.0 |
| concepts[11].wikidata | https://www.wikidata.org/wiki/Q234881 |
| concepts[11].display_name | Quantum electrodynamics |
| concepts[12].id | https://openalex.org/C153294291 |
| concepts[12].level | 1 |
| concepts[12].score | 0.0 |
| concepts[12].wikidata | https://www.wikidata.org/wiki/Q25261 |
| concepts[12].display_name | Meteorology |
| concepts[13].id | https://openalex.org/C121332964 |
| concepts[13].level | 0 |
| concepts[13].score | 0.0 |
| concepts[13].wikidata | https://www.wikidata.org/wiki/Q413 |
| concepts[13].display_name | Physics |
| keywords[0].id | https://openalex.org/keywords/training-set |
| keywords[0].score | 0.7339940071105957 |
| keywords[0].display_name | Training set |
| keywords[1].id | https://openalex.org/keywords/labeled-data |
| keywords[1].score | 0.7283814549446106 |
| keywords[1].display_name | Labeled data |
| keywords[2].id | https://openalex.org/keywords/computer-science |
| keywords[2].score | 0.723684549331665 |
| keywords[2].display_name | Computer science |
| keywords[3].id | https://openalex.org/keywords/pipeline |
| keywords[3].score | 0.6110484004020691 |
| keywords[3].display_name | Pipeline (software) |
| keywords[4].id | https://openalex.org/keywords/artificial-intelligence |
| keywords[4].score | 0.594964325428009 |
| keywords[4].display_name | Artificial intelligence |
| keywords[5].id | https://openalex.org/keywords/training |
| keywords[5].score | 0.5541278719902039 |
| keywords[5].display_name | Training (meteorology) |
| keywords[6].id | https://openalex.org/keywords/machine-learning |
| keywords[6].score | 0.5389244556427002 |
| keywords[6].display_name | Machine learning |
| keywords[7].id | https://openalex.org/keywords/set |
| keywords[7].score | 0.5050604939460754 |
| keywords[7].display_name | Set (abstract data type) |
| keywords[8].id | https://openalex.org/keywords/data-set |
| keywords[8].score | 0.4662512540817261 |
| keywords[8].display_name | Data set |
| keywords[9].id | https://openalex.org/keywords/self-consistent |
| keywords[9].score | 0.41620492935180664 |
| keywords[9].display_name | Self consistent |
| language | en |
| locations[0].id | pmh:oai:arXiv.org:2301.07294 |
| 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/2301.07294 |
| locations[0].version | submittedVersion |
| locations[0].raw_type | |
| 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/2301.07294 |
| locations[1].id | doi:10.48550/arxiv.2301.07294 |
| 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.2301.07294 |
| indexed_in | arxiv, datacite |
| authorships[0].author.id | https://openalex.org/A5010545980 |
| authorships[0].author.orcid | |
| authorships[0].author.display_name | Aswathnarayan Radhakrishnan |
| authorships[0].author_position | first |
| authorships[0].raw_author_name | Radhakrishnan, Aswathnarayan |
| authorships[0].is_corresponding | False |
| authorships[1].author.id | https://openalex.org/A5012185466 |
| authorships[1].author.orcid | https://orcid.org/0000-0002-3364-027X |
| authorships[1].author.display_name | Jim Davis |
| authorships[1].author_position | middle |
| authorships[1].raw_author_name | Davis, Jim |
| authorships[1].is_corresponding | False |
| authorships[2].author.id | https://openalex.org/A5065643938 |
| authorships[2].author.orcid | |
| authorships[2].author.display_name | Zachary Rabin |
| authorships[2].author_position | middle |
| authorships[2].raw_author_name | Rabin, Zachary |
| authorships[2].is_corresponding | False |
| authorships[3].author.id | https://openalex.org/A5062874503 |
| authorships[3].author.orcid | https://orcid.org/0000-0002-4135-3967 |
| authorships[3].author.display_name | Benjamin M. Lewis |
| authorships[3].author_position | middle |
| authorships[3].raw_author_name | Lewis, Benjamin |
| authorships[3].is_corresponding | False |
| authorships[4].author.id | https://openalex.org/A5039748817 |
| authorships[4].author.orcid | https://orcid.org/0000-0003-2575-0281 |
| authorships[4].author.display_name | Matthew Scherreik |
| authorships[4].author_position | middle |
| authorships[4].raw_author_name | Scherreik, Matthew |
| authorships[4].is_corresponding | False |
| authorships[5].author.id | https://openalex.org/A5022912549 |
| authorships[5].author.orcid | https://orcid.org/0000-0003-4922-2598 |
| authorships[5].author.display_name | Roman Ilin |
| authorships[5].author_position | last |
| authorships[5].raw_author_name | Ilin, Roman |
| 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/2301.07294 |
| open_access.oa_status | green |
| open_access.any_repository_has_fulltext | False |
| created_date | 2025-10-10T00:00:00 |
| display_name | Enhancing Self-Training Methods |
| has_fulltext | False |
| is_retracted | False |
| updated_date | 2025-11-06T06:51:31.235846 |
| primary_topic.id | https://openalex.org/T12535 |
| primary_topic.field.id | https://openalex.org/fields/17 |
| primary_topic.field.display_name | Computer Science |
| primary_topic.score | 0.9986000061035156 |
| 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 | Machine Learning and Data Classification |
| related_works | https://openalex.org/W2161052216, https://openalex.org/W2130553454, https://openalex.org/W4317548404, https://openalex.org/W2786391746, https://openalex.org/W3022007134, https://openalex.org/W2087783760, https://openalex.org/W2033364610, https://openalex.org/W2797776314, https://openalex.org/W3163689946, https://openalex.org/W2153927146 |
| cited_by_count | 0 |
| locations_count | 2 |
| best_oa_location.id | pmh:oai:arXiv.org:2301.07294 |
| 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/2301.07294 |
| best_oa_location.version | submittedVersion |
| best_oa_location.raw_type | |
| 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/2301.07294 |
| primary_location.id | pmh:oai:arXiv.org:2301.07294 |
| 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/2301.07294 |
| primary_location.version | submittedVersion |
| primary_location.raw_type | |
| 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/2301.07294 |
| publication_date | 2023-01-18 |
| publication_year | 2023 |
| referenced_works_count | 0 |
| abstract_inverted_index.a | 19, 71 |
| abstract_inverted_index.In | 74 |
| abstract_inverted_index.We | 93 |
| abstract_inverted_index.by | 44 |
| abstract_inverted_index.in | 56, 66, 127 |
| abstract_inverted_index.is | 18 |
| abstract_inverted_index.of | 7, 14, 29, 90, 114 |
| abstract_inverted_index.on | 4 |
| abstract_inverted_index.to | 40, 63, 81, 86, 118 |
| abstract_inverted_index.we | 77, 109 |
| abstract_inverted_index.Set | 120 |
| abstract_inverted_index.few | 72 |
| abstract_inverted_index.for | 48 |
| abstract_inverted_index.not | 125 |
| abstract_inverted_index.our | 95, 115 |
| abstract_inverted_index.the | 27, 35, 45, 49, 83, 88, 112 |
| abstract_inverted_index.Open | 119 |
| abstract_inverted_index.This | 52 |
| abstract_inverted_index.also | 110 |
| abstract_inverted_index.bias | 53 |
| abstract_inverted_index.data | 9, 122 |
| abstract_inverted_index.from | 26 |
| abstract_inverted_index.just | 70 |
| abstract_inverted_index.over | 97, 103 |
| abstract_inverted_index.seen | 126 |
| abstract_inverted_index.sets | 6, 13 |
| abstract_inverted_index.that | 23, 32 |
| abstract_inverted_index.this | 75 |
| abstract_inverted_index.when | 34 |
| abstract_inverted_index.with | 11 |
| abstract_inverted_index.after | 69 |
| abstract_inverted_index.along | 10 |
| abstract_inverted_index.bias" | 31 |
| abstract_inverted_index.bias. | 92 |
| abstract_inverted_index.data. | 16, 51 |
| abstract_inverted_index.gains | 102 |
| abstract_inverted_index.given | 43 |
| abstract_inverted_index.large | 12 |
| abstract_inverted_index.model | 37, 47, 67 |
| abstract_inverted_index.often | 24 |
| abstract_inverted_index.small | 5 |
| abstract_inverted_index.study | 111 |
| abstract_inverted_index.train | 3 |
| abstract_inverted_index.work, | 76 |
| abstract_inverted_index.across | 59 |
| abstract_inverted_index.data). | 129 |
| abstract_inverted_index.design | 106 |
| abstract_inverted_index.effect | 89 |
| abstract_inverted_index.occurs | 33 |
| abstract_inverted_index.classes | 124 |
| abstract_inverted_index.impedes | 54 |
| abstract_inverted_index.improve | 82 |
| abstract_inverted_index.labeled | 8, 128 |
| abstract_inverted_index.leading | 62 |
| abstract_inverted_index.problem | 28 |
| abstract_inverted_index.showing | 100 |
| abstract_inverted_index.student | 36 |
| abstract_inverted_index.suffers | 25 |
| abstract_inverted_index.teacher | 46 |
| abstract_inverted_index.Finally, | 108 |
| abstract_inverted_index.accuracy | 58 |
| abstract_inverted_index.approach | 22, 117 |
| abstract_inverted_index.choices. | 107 |
| abstract_inverted_index.datasets | 99 |
| abstract_inverted_index.describe | 78 |
| abstract_inverted_index.enhanced | 116 |
| abstract_inverted_index.evaluate | 94 |
| abstract_inverted_index.existing | 104 |
| abstract_inverted_index.learning | 1 |
| abstract_inverted_index.mitigate | 87 |
| abstract_inverted_index.multiple | 79, 98 |
| abstract_inverted_index.overfits | 39 |
| abstract_inverted_index.pipeline | 85 |
| abstract_inverted_index.unwanted | 64 |
| abstract_inverted_index.incorrect | 41 |
| abstract_inverted_index.unlabeled | 15, 50, 121 |
| abstract_inverted_index.approaches | 2 |
| abstract_inverted_index.repeatedly | 38 |
| abstract_inverted_index.saturation | 65 |
| abstract_inverted_index.(containing | 123 |
| abstract_inverted_index.iterations, | 61 |
| abstract_inverted_index.iterations. | 73 |
| abstract_inverted_index.performance | 68, 101 |
| abstract_inverted_index.confirmation | 91 |
| abstract_inverted_index.enhancements | 80, 96 |
| abstract_inverted_index.improvements | 55 |
| abstract_inverted_index.pseudo-label | 57 |
| abstract_inverted_index."confirmation | 30 |
| abstract_inverted_index.Self-training | 17 |
| abstract_inverted_index.extendability | 113 |
| abstract_inverted_index.pseudo-labels | 42 |
| abstract_inverted_index.self-training | 60, 84, 105 |
| abstract_inverted_index.Semi-supervised | 0 |
| abstract_inverted_index.semi-supervised | 20 |
| abstract_inverted_index.teacher-student | 21 |
| cited_by_percentile_year | |
| countries_distinct_count | 0 |
| institutions_distinct_count | 6 |
| sustainable_development_goals[0].id | https://metadata.un.org/sdg/4 |
| sustainable_development_goals[0].score | 0.5099999904632568 |
| sustainable_development_goals[0].display_name | Quality Education |
| citation_normalized_percentile |