Fair Classifiers that Abstain without Harm Article Swipe
Tongxin Yin
,
Jean-François Ton
,
Ruocheng Guo
,
Yuanshun Yao
,
Mingyan Liu
,
Yang Liu
·
YOU?
·
· 2023
· Open Access
·
· DOI: https://doi.org/10.48550/arxiv.2310.06205
YOU?
·
· 2023
· Open Access
·
· DOI: https://doi.org/10.48550/arxiv.2310.06205
In critical applications, it is vital for classifiers to defer decision-making to humans. We propose a post-hoc method that makes existing classifiers selectively abstain from predicting certain samples. Our abstaining classifier is incentivized to maintain the original accuracy for each sub-population (i.e. no harm) while achieving a set of group fairness definitions to a user specified degree. To this end, we design an Integer Programming (IP) procedure that assigns abstention decisions for each training sample to satisfy a set of constraints. To generalize the abstaining decisions to test samples, we then train a surrogate model to learn the abstaining decisions based on the IP solutions in an end-to-end manner. We analyze the feasibility of the IP procedure to determine the possible abstention rate for different levels of unfairness tolerance and accuracy constraint for achieving no harm. To the best of our knowledge, this work is the first to identify the theoretical relationships between the constraint parameters and the required abstention rate. Our theoretical results are important since a high abstention rate is often infeasible in practice due to a lack of human resources. Our framework outperforms existing methods in terms of fairness disparity without sacrificing accuracy at similar abstention rates.
Related Topics
Concepts
Metadata
- Type
- preprint
- Language
- en
- Landing Page
- http://arxiv.org/abs/2310.06205
- https://arxiv.org/pdf/2310.06205
- OA Status
- green
- Related Works
- 10
- OpenAlex ID
- https://openalex.org/W4387560835
All OpenAlex metadata
Raw OpenAlex JSON
- OpenAlex ID
-
https://openalex.org/W4387560835Canonical identifier for this work in OpenAlex
- DOI
-
https://doi.org/10.48550/arxiv.2310.06205Digital Object Identifier
- Title
-
Fair Classifiers that Abstain without HarmWork title
- Type
-
preprintOpenAlex work type
- Language
-
enPrimary language
- Publication year
-
2023Year of publication
- Publication date
-
2023-10-09Full publication date if available
- Authors
-
Tongxin Yin, Jean-François Ton, Ruocheng Guo, Yuanshun Yao, Mingyan Liu, Yang LiuList of authors in order
- Landing page
-
https://arxiv.org/abs/2310.06205Publisher landing page
- PDF URL
-
https://arxiv.org/pdf/2310.06205Direct 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/2310.06205Direct OA link when available
- Concepts
-
Harm, Computer science, Classifier (UML), Constraint (computer-aided design), Artificial intelligence, Set (abstract data type), Machine learning, Mathematics, Psychology, Social psychology, Geometry, 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/W4387560835 |
|---|---|
| doi | https://doi.org/10.48550/arxiv.2310.06205 |
| ids.doi | https://doi.org/10.48550/arxiv.2310.06205 |
| ids.openalex | https://openalex.org/W4387560835 |
| fwci | |
| type | preprint |
| title | Fair Classifiers that Abstain without Harm |
| biblio.issue | |
| biblio.volume | |
| biblio.last_page | |
| biblio.first_page | |
| topics[0].id | https://openalex.org/T11689 |
| topics[0].field.id | https://openalex.org/fields/17 |
| topics[0].field.display_name | Computer Science |
| topics[0].score | 0.9747999906539917 |
| 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 | Adversarial Robustness in Machine Learning |
| topics[1].id | https://openalex.org/T10883 |
| topics[1].field.id | https://openalex.org/fields/33 |
| topics[1].field.display_name | Social Sciences |
| topics[1].score | 0.9671000242233276 |
| topics[1].domain.id | https://openalex.org/domains/2 |
| topics[1].domain.display_name | Social Sciences |
| topics[1].subfield.id | https://openalex.org/subfields/3311 |
| topics[1].subfield.display_name | Safety Research |
| topics[1].display_name | Ethics and Social Impacts of AI |
| topics[2].id | https://openalex.org/T11997 |
| topics[2].field.id | https://openalex.org/fields/28 |
| topics[2].field.display_name | Neuroscience |
| topics[2].score | 0.9056000113487244 |
| topics[2].domain.id | https://openalex.org/domains/1 |
| topics[2].domain.display_name | Life Sciences |
| topics[2].subfield.id | https://openalex.org/subfields/2805 |
| topics[2].subfield.display_name | Cognitive Neuroscience |
| topics[2].display_name | Free Will and Agency |
| is_xpac | False |
| apc_list | |
| apc_paid | |
| concepts[0].id | https://openalex.org/C2777363581 |
| concepts[0].level | 2 |
| concepts[0].score | 0.7471894025802612 |
| concepts[0].wikidata | https://www.wikidata.org/wiki/Q15098235 |
| concepts[0].display_name | Harm |
| concepts[1].id | https://openalex.org/C41008148 |
| concepts[1].level | 0 |
| concepts[1].score | 0.6624596118927002 |
| concepts[1].wikidata | https://www.wikidata.org/wiki/Q21198 |
| concepts[1].display_name | Computer science |
| concepts[2].id | https://openalex.org/C95623464 |
| concepts[2].level | 2 |
| concepts[2].score | 0.5551252961158752 |
| concepts[2].wikidata | https://www.wikidata.org/wiki/Q1096149 |
| concepts[2].display_name | Classifier (UML) |
| concepts[3].id | https://openalex.org/C2776036281 |
| concepts[3].level | 2 |
| concepts[3].score | 0.5198312401771545 |
| concepts[3].wikidata | https://www.wikidata.org/wiki/Q48769818 |
| concepts[3].display_name | Constraint (computer-aided design) |
| concepts[4].id | https://openalex.org/C154945302 |
| concepts[4].level | 1 |
| concepts[4].score | 0.46662580966949463 |
| concepts[4].wikidata | https://www.wikidata.org/wiki/Q11660 |
| concepts[4].display_name | Artificial intelligence |
| concepts[5].id | https://openalex.org/C177264268 |
| concepts[5].level | 2 |
| concepts[5].score | 0.4330371618270874 |
| concepts[5].wikidata | https://www.wikidata.org/wiki/Q1514741 |
| concepts[5].display_name | Set (abstract data type) |
| concepts[6].id | https://openalex.org/C119857082 |
| concepts[6].level | 1 |
| concepts[6].score | 0.38413485884666443 |
| concepts[6].wikidata | https://www.wikidata.org/wiki/Q2539 |
| concepts[6].display_name | Machine learning |
| concepts[7].id | https://openalex.org/C33923547 |
| concepts[7].level | 0 |
| concepts[7].score | 0.17019003629684448 |
| concepts[7].wikidata | https://www.wikidata.org/wiki/Q395 |
| concepts[7].display_name | Mathematics |
| concepts[8].id | https://openalex.org/C15744967 |
| concepts[8].level | 0 |
| concepts[8].score | 0.10871565341949463 |
| concepts[8].wikidata | https://www.wikidata.org/wiki/Q9418 |
| concepts[8].display_name | Psychology |
| concepts[9].id | https://openalex.org/C77805123 |
| concepts[9].level | 1 |
| concepts[9].score | 0.07748827338218689 |
| concepts[9].wikidata | https://www.wikidata.org/wiki/Q161272 |
| concepts[9].display_name | Social psychology |
| concepts[10].id | https://openalex.org/C2524010 |
| concepts[10].level | 1 |
| concepts[10].score | 0.0 |
| concepts[10].wikidata | https://www.wikidata.org/wiki/Q8087 |
| concepts[10].display_name | Geometry |
| concepts[11].id | https://openalex.org/C199360897 |
| concepts[11].level | 1 |
| concepts[11].score | 0.0 |
| concepts[11].wikidata | https://www.wikidata.org/wiki/Q9143 |
| concepts[11].display_name | Programming language |
| keywords[0].id | https://openalex.org/keywords/harm |
| keywords[0].score | 0.7471894025802612 |
| keywords[0].display_name | Harm |
| keywords[1].id | https://openalex.org/keywords/computer-science |
| keywords[1].score | 0.6624596118927002 |
| keywords[1].display_name | Computer science |
| keywords[2].id | https://openalex.org/keywords/classifier |
| keywords[2].score | 0.5551252961158752 |
| keywords[2].display_name | Classifier (UML) |
| keywords[3].id | https://openalex.org/keywords/constraint |
| keywords[3].score | 0.5198312401771545 |
| keywords[3].display_name | Constraint (computer-aided design) |
| keywords[4].id | https://openalex.org/keywords/artificial-intelligence |
| keywords[4].score | 0.46662580966949463 |
| keywords[4].display_name | Artificial intelligence |
| keywords[5].id | https://openalex.org/keywords/set |
| keywords[5].score | 0.4330371618270874 |
| keywords[5].display_name | Set (abstract data type) |
| keywords[6].id | https://openalex.org/keywords/machine-learning |
| keywords[6].score | 0.38413485884666443 |
| keywords[6].display_name | Machine learning |
| keywords[7].id | https://openalex.org/keywords/mathematics |
| keywords[7].score | 0.17019003629684448 |
| keywords[7].display_name | Mathematics |
| keywords[8].id | https://openalex.org/keywords/psychology |
| keywords[8].score | 0.10871565341949463 |
| keywords[8].display_name | Psychology |
| keywords[9].id | https://openalex.org/keywords/social-psychology |
| keywords[9].score | 0.07748827338218689 |
| keywords[9].display_name | Social psychology |
| language | en |
| locations[0].id | pmh:oai:arXiv.org:2310.06205 |
| 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/2310.06205 |
| 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/2310.06205 |
| locations[1].id | doi:10.48550/arxiv.2310.06205 |
| 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 | |
| locations[1].pdf_url | |
| locations[1].version | |
| locations[1].raw_type | article |
| locations[1].license_id | |
| 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.2310.06205 |
| indexed_in | arxiv, datacite |
| authorships[0].author.id | https://openalex.org/A5112909068 |
| authorships[0].author.orcid | |
| authorships[0].author.display_name | Tongxin Yin |
| authorships[0].author_position | first |
| authorships[0].raw_author_name | Yin, Tongxin |
| authorships[0].is_corresponding | False |
| authorships[1].author.id | https://openalex.org/A5069983126 |
| authorships[1].author.orcid | https://orcid.org/0000-0002-2756-8945 |
| authorships[1].author.display_name | Jean-François Ton |
| authorships[1].author_position | middle |
| authorships[1].raw_author_name | Ton, Jean-François |
| authorships[1].is_corresponding | False |
| authorships[2].author.id | https://openalex.org/A5054719216 |
| authorships[2].author.orcid | https://orcid.org/0000-0002-8522-6142 |
| authorships[2].author.display_name | Ruocheng Guo |
| authorships[2].author_position | middle |
| authorships[2].raw_author_name | Guo, Ruocheng |
| authorships[2].is_corresponding | False |
| authorships[3].author.id | https://openalex.org/A5014146843 |
| authorships[3].author.orcid | |
| authorships[3].author.display_name | Yuanshun Yao |
| authorships[3].author_position | middle |
| authorships[3].raw_author_name | Yao, Yuanshun |
| authorships[3].is_corresponding | False |
| authorships[4].author.id | https://openalex.org/A5101967011 |
| authorships[4].author.orcid | https://orcid.org/0000-0003-3295-9200 |
| authorships[4].author.display_name | Mingyan Liu |
| authorships[4].author_position | middle |
| authorships[4].raw_author_name | Liu, Mingyan |
| authorships[4].is_corresponding | False |
| authorships[5].author.id | https://openalex.org/A5100355692 |
| authorships[5].author.orcid | https://orcid.org/0000-0001-7300-9215 |
| authorships[5].author.display_name | Yang Liu |
| authorships[5].author_position | last |
| authorships[5].raw_author_name | Liu, Yang |
| 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/2310.06205 |
| open_access.oa_status | green |
| open_access.any_repository_has_fulltext | False |
| created_date | 2025-10-10T00:00:00 |
| display_name | Fair Classifiers that Abstain without Harm |
| has_fulltext | False |
| is_retracted | False |
| updated_date | 2025-11-06T06:51:31.235846 |
| primary_topic.id | https://openalex.org/T11689 |
| primary_topic.field.id | https://openalex.org/fields/17 |
| primary_topic.field.display_name | Computer Science |
| primary_topic.score | 0.9747999906539917 |
| 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 | Adversarial Robustness in Machine Learning |
| related_works | https://openalex.org/W2356901839, https://openalex.org/W3203175338, https://openalex.org/W3209501579, https://openalex.org/W2969547062, https://openalex.org/W2497114785, https://openalex.org/W4283162910, https://openalex.org/W2803806723, https://openalex.org/W119609074, https://openalex.org/W4245578471, https://openalex.org/W4293754511 |
| cited_by_count | 0 |
| locations_count | 2 |
| best_oa_location.id | pmh:oai:arXiv.org:2310.06205 |
| 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/2310.06205 |
| 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/2310.06205 |
| primary_location.id | pmh:oai:arXiv.org:2310.06205 |
| 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/2310.06205 |
| 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/2310.06205 |
| publication_date | 2023-10-09 |
| publication_year | 2023 |
| referenced_works_count | 0 |
| abstract_inverted_index.a | 15, 46, 53, 77, 92, 167, 178 |
| abstract_inverted_index.IP | 103, 115 |
| abstract_inverted_index.In | 0 |
| abstract_inverted_index.To | 57, 81, 136 |
| abstract_inverted_index.We | 13, 109 |
| abstract_inverted_index.an | 62, 106 |
| abstract_inverted_index.at | 196 |
| abstract_inverted_index.in | 105, 174, 188 |
| abstract_inverted_index.is | 4, 31, 144, 171 |
| abstract_inverted_index.it | 3 |
| abstract_inverted_index.no | 42, 134 |
| abstract_inverted_index.of | 48, 79, 113, 126, 139, 180, 190 |
| abstract_inverted_index.on | 101 |
| abstract_inverted_index.to | 8, 11, 33, 52, 75, 86, 95, 117, 147, 177 |
| abstract_inverted_index.we | 60, 89 |
| abstract_inverted_index.Our | 28, 161, 183 |
| abstract_inverted_index.and | 129, 156 |
| abstract_inverted_index.are | 164 |
| abstract_inverted_index.due | 176 |
| abstract_inverted_index.for | 6, 38, 71, 123, 132 |
| abstract_inverted_index.our | 140 |
| abstract_inverted_index.set | 47, 78 |
| abstract_inverted_index.the | 35, 83, 97, 102, 111, 114, 119, 137, 145, 149, 153, 157 |
| abstract_inverted_index.(IP) | 65 |
| abstract_inverted_index.best | 138 |
| abstract_inverted_index.each | 39, 72 |
| abstract_inverted_index.end, | 59 |
| abstract_inverted_index.from | 24 |
| abstract_inverted_index.high | 168 |
| abstract_inverted_index.lack | 179 |
| abstract_inverted_index.rate | 122, 170 |
| abstract_inverted_index.test | 87 |
| abstract_inverted_index.that | 18, 67 |
| abstract_inverted_index.then | 90 |
| abstract_inverted_index.this | 58, 142 |
| abstract_inverted_index.user | 54 |
| abstract_inverted_index.work | 143 |
| abstract_inverted_index.(i.e. | 41 |
| abstract_inverted_index.based | 100 |
| abstract_inverted_index.defer | 9 |
| abstract_inverted_index.first | 146 |
| abstract_inverted_index.group | 49 |
| abstract_inverted_index.harm) | 43 |
| abstract_inverted_index.harm. | 135 |
| abstract_inverted_index.human | 181 |
| abstract_inverted_index.learn | 96 |
| abstract_inverted_index.makes | 19 |
| abstract_inverted_index.model | 94 |
| abstract_inverted_index.often | 172 |
| abstract_inverted_index.rate. | 160 |
| abstract_inverted_index.since | 166 |
| abstract_inverted_index.terms | 189 |
| abstract_inverted_index.train | 91 |
| abstract_inverted_index.vital | 5 |
| abstract_inverted_index.while | 44 |
| abstract_inverted_index.design | 61 |
| abstract_inverted_index.levels | 125 |
| abstract_inverted_index.method | 17 |
| abstract_inverted_index.rates. | 199 |
| abstract_inverted_index.sample | 74 |
| abstract_inverted_index.Integer | 63 |
| abstract_inverted_index.abstain | 23 |
| abstract_inverted_index.analyze | 110 |
| abstract_inverted_index.assigns | 68 |
| abstract_inverted_index.between | 152 |
| abstract_inverted_index.certain | 26 |
| abstract_inverted_index.degree. | 56 |
| abstract_inverted_index.humans. | 12 |
| abstract_inverted_index.manner. | 108 |
| abstract_inverted_index.methods | 187 |
| abstract_inverted_index.propose | 14 |
| abstract_inverted_index.results | 163 |
| abstract_inverted_index.satisfy | 76 |
| abstract_inverted_index.similar | 197 |
| abstract_inverted_index.without | 193 |
| abstract_inverted_index.accuracy | 37, 130, 195 |
| abstract_inverted_index.critical | 1 |
| abstract_inverted_index.existing | 20, 186 |
| abstract_inverted_index.fairness | 50, 191 |
| abstract_inverted_index.identify | 148 |
| abstract_inverted_index.maintain | 34 |
| abstract_inverted_index.original | 36 |
| abstract_inverted_index.possible | 120 |
| abstract_inverted_index.post-hoc | 16 |
| abstract_inverted_index.practice | 175 |
| abstract_inverted_index.required | 158 |
| abstract_inverted_index.samples, | 88 |
| abstract_inverted_index.samples. | 27 |
| abstract_inverted_index.training | 73 |
| abstract_inverted_index.achieving | 45, 133 |
| abstract_inverted_index.decisions | 70, 85, 99 |
| abstract_inverted_index.determine | 118 |
| abstract_inverted_index.different | 124 |
| abstract_inverted_index.disparity | 192 |
| abstract_inverted_index.framework | 184 |
| abstract_inverted_index.important | 165 |
| abstract_inverted_index.procedure | 66, 116 |
| abstract_inverted_index.solutions | 104 |
| abstract_inverted_index.specified | 55 |
| abstract_inverted_index.surrogate | 93 |
| abstract_inverted_index.tolerance | 128 |
| abstract_inverted_index.abstaining | 29, 84, 98 |
| abstract_inverted_index.abstention | 69, 121, 159, 169, 198 |
| abstract_inverted_index.classifier | 30 |
| abstract_inverted_index.constraint | 131, 154 |
| abstract_inverted_index.end-to-end | 107 |
| abstract_inverted_index.generalize | 82 |
| abstract_inverted_index.infeasible | 173 |
| abstract_inverted_index.knowledge, | 141 |
| abstract_inverted_index.parameters | 155 |
| abstract_inverted_index.predicting | 25 |
| abstract_inverted_index.resources. | 182 |
| abstract_inverted_index.unfairness | 127 |
| abstract_inverted_index.Programming | 64 |
| abstract_inverted_index.classifiers | 7, 21 |
| abstract_inverted_index.definitions | 51 |
| abstract_inverted_index.feasibility | 112 |
| abstract_inverted_index.outperforms | 185 |
| abstract_inverted_index.sacrificing | 194 |
| abstract_inverted_index.selectively | 22 |
| abstract_inverted_index.theoretical | 150, 162 |
| abstract_inverted_index.constraints. | 80 |
| abstract_inverted_index.incentivized | 32 |
| abstract_inverted_index.applications, | 2 |
| abstract_inverted_index.relationships | 151 |
| abstract_inverted_index.sub-population | 40 |
| abstract_inverted_index.decision-making | 10 |
| cited_by_percentile_year | |
| countries_distinct_count | 0 |
| institutions_distinct_count | 6 |
| sustainable_development_goals[0].id | https://metadata.un.org/sdg/16 |
| sustainable_development_goals[0].score | 0.8399999737739563 |
| sustainable_development_goals[0].display_name | Peace, Justice and strong institutions |
| citation_normalized_percentile |