Equal Opportunity of Coverage in Fair Regression Article Swipe

PDF

Fangxin Wang , Lu Cheng , Ruocheng Guo , Kay Liu , Philip S. Yu ·

YOU? · · 2023 · Open Access · · DOI: https://doi.org/10.48550/arxiv.2311.02243

We study fair machine learning (ML) under predictive uncertainty to enable reliable and trustworthy decision-making. The seminal work of ``equalized coverage'' proposed an uncertainty-aware fairness notion. However, it does not guarantee equal coverage rates across more fine-grained groups (e.g., low-income females) conditioning on the true label and is biased in the assessment of uncertainty. To tackle these limitations, we propose a new uncertainty-aware fairness -- Equal Opportunity of Coverage (EOC) -- that aims to achieve two properties: (1) coverage rates for different groups with similar outcomes are close, and (2) the coverage rate for the entire population remains at a predetermined level. Further, the prediction intervals should be narrow to be informative. We propose Binned Fair Quantile Regression (BFQR), a distribution-free post-processing method to improve EOC with reasonable width for any trained ML models. It first calibrates a hold-out set to bound deviation from EOC, then leverages conformal prediction to maintain EOC on a test set, meanwhile optimizing prediction interval width. Experimental results demonstrate the effectiveness of our method in improving EOC. Our code is publicly available at https://github.com/fangxin-wang/bfqr .

Related Topics

Artificial Intelligence

Concepts

Computer science Quantile regression Set (abstract data type) Coverage probability Trustworthiness Interval (graph theory) Regression Quantile Population Code (set theory) Machine learning Statistics Confidence interval Artificial intelligence Econometrics Mathematics Computer security Combinatorics Sociology Demography Programming language

Metadata

Type: preprint
Language: en
Landing Page: http://arxiv.org/abs/2311.02243
PDF: https://arxiv.org/pdf/2311.02243
OA Status: green
Cited By: 3
Related Works: 10
OpenAlex ID: https://openalex.org/W4388512345

All OpenAlex metadata

Raw OpenAlex JSON

OpenAlex ID: https://openalex.org/W4388512345

Canonical identifier for this work in OpenAlex
DOI: https://doi.org/10.48550/arxiv.2311.02243

Digital Object Identifier
Title: Equal Opportunity of Coverage in Fair Regression

Work title
Type: preprint

OpenAlex work type
Language: en

Primary language
Publication year: 2023

Year of publication
Publication date: 2023-11-03

Full publication date if available
Authors: Fangxin Wang, Lu Cheng, Ruocheng Guo, Kay Liu, Philip S. Yu

List of authors in order
Landing page: https://arxiv.org/abs/2311.02243

Publisher landing page
PDF URL: https://arxiv.org/pdf/2311.02243

Direct link to full text PDF
Open access: Yes

Whether a free full text is available
OA status: green

Open access status per OpenAlex
OA URL: https://arxiv.org/pdf/2311.02243

Direct OA link when available
Concepts: Computer science, Quantile regression, Set (abstract data type), Coverage probability, Trustworthiness, Interval (graph theory), Regression, Quantile, Population, Code (set theory), Machine learning, Statistics, Confidence interval, Artificial intelligence, Econometrics, Mathematics, Computer security, Combinatorics, Sociology, Demography, Programming language

Top concepts (fields/topics) attached by OpenAlex
Cited by: 3

Total citation count in OpenAlex
Citations by year (recent): 2024: 1, 2023: 2

Per-year citation counts (last 5 years)
Related works (count): 10

Other works algorithmically related by OpenAlex

Full payload

id	https://openalex.org/W4388512345
doi	https://doi.org/10.48550/arxiv.2311.02243
ids.doi	https://doi.org/10.48550/arxiv.2311.02243
ids.openalex	https://openalex.org/W4388512345
fwci
type	preprint
title	Equal Opportunity of Coverage in Fair Regression
biblio.issue
biblio.volume
biblio.last_page
biblio.first_page
topics[0].id	https://openalex.org/T10883
topics[0].field.id	https://openalex.org/fields/33
topics[0].field.display_name	Social Sciences
topics[0].score	0.9821000099182129
topics[0].domain.id	https://openalex.org/domains/2
topics[0].domain.display_name	Social Sciences
topics[0].subfield.id	https://openalex.org/subfields/3311
topics[0].subfield.display_name	Safety Research
topics[0].display_name	Ethics and Social Impacts of AI
is_xpac	False
apc_list
apc_paid
concepts[0].id	https://openalex.org/C41008148
concepts[0].level	0
concepts[0].score	0.6417620182037354
concepts[0].wikidata	https://www.wikidata.org/wiki/Q21198
concepts[0].display_name	Computer science
concepts[1].id	https://openalex.org/C63817138
concepts[1].level	2
concepts[1].score	0.5888808369636536
concepts[1].wikidata	https://www.wikidata.org/wiki/Q3455889
concepts[1].display_name	Quantile regression
concepts[2].id	https://openalex.org/C177264268
concepts[2].level	2
concepts[2].score	0.5677388310432434
concepts[2].wikidata	https://www.wikidata.org/wiki/Q1514741
concepts[2].display_name	Set (abstract data type)
concepts[3].id	https://openalex.org/C2776292839
concepts[3].level	3
concepts[3].score	0.520834743976593
concepts[3].wikidata	https://www.wikidata.org/wiki/Q5179217
concepts[3].display_name	Coverage probability
concepts[4].id	https://openalex.org/C153701036
concepts[4].level	2
concepts[4].score	0.4965875744819641
concepts[4].wikidata	https://www.wikidata.org/wiki/Q659974
concepts[4].display_name	Trustworthiness
concepts[5].id	https://openalex.org/C2778067643
concepts[5].level	2
concepts[5].score	0.4946124255657196
concepts[5].wikidata	https://www.wikidata.org/wiki/Q166507
concepts[5].display_name	Interval (graph theory)
concepts[6].id	https://openalex.org/C83546350
concepts[6].level	2
concepts[6].score	0.479521781206131
concepts[6].wikidata	https://www.wikidata.org/wiki/Q1139051
concepts[6].display_name	Regression
concepts[7].id	https://openalex.org/C118671147
concepts[7].level	2
concepts[7].score	0.46922004222869873
concepts[7].wikidata	https://www.wikidata.org/wiki/Q578714
concepts[7].display_name	Quantile
concepts[8].id	https://openalex.org/C2908647359
concepts[8].level	2
concepts[8].score	0.4299023747444153
concepts[8].wikidata	https://www.wikidata.org/wiki/Q2625603
concepts[8].display_name	Population
concepts[9].id	https://openalex.org/C2776760102
concepts[9].level	3
concepts[9].score	0.42871496081352234
concepts[9].wikidata	https://www.wikidata.org/wiki/Q5139990
concepts[9].display_name	Code (set theory)
concepts[10].id	https://openalex.org/C119857082
concepts[10].level	1
concepts[10].score	0.38563358783721924
concepts[10].wikidata	https://www.wikidata.org/wiki/Q2539
concepts[10].display_name	Machine learning
concepts[11].id	https://openalex.org/C105795698
concepts[11].level	1
concepts[11].score	0.35695305466651917
concepts[11].wikidata	https://www.wikidata.org/wiki/Q12483
concepts[11].display_name	Statistics
concepts[12].id	https://openalex.org/C44249647
concepts[12].level	2
concepts[12].score	0.34881192445755005
concepts[12].wikidata	https://www.wikidata.org/wiki/Q208498
concepts[12].display_name	Confidence interval
concepts[13].id	https://openalex.org/C154945302
concepts[13].level	1
concepts[13].score	0.3415707051753998
concepts[13].wikidata	https://www.wikidata.org/wiki/Q11660
concepts[13].display_name	Artificial intelligence
concepts[14].id	https://openalex.org/C149782125
concepts[14].level	1
concepts[14].score	0.32040315866470337
concepts[14].wikidata	https://www.wikidata.org/wiki/Q160039
concepts[14].display_name	Econometrics
concepts[15].id	https://openalex.org/C33923547
concepts[15].level	0
concepts[15].score	0.2153456211090088
concepts[15].wikidata	https://www.wikidata.org/wiki/Q395
concepts[15].display_name	Mathematics
concepts[16].id	https://openalex.org/C38652104
concepts[16].level	1
concepts[16].score	0.07599031925201416
concepts[16].wikidata	https://www.wikidata.org/wiki/Q3510521
concepts[16].display_name	Computer security
concepts[17].id	https://openalex.org/C114614502
concepts[17].level	1
concepts[17].score	0.0
concepts[17].wikidata	https://www.wikidata.org/wiki/Q76592
concepts[17].display_name	Combinatorics
concepts[18].id	https://openalex.org/C144024400
concepts[18].level	0
concepts[18].score	0.0
concepts[18].wikidata	https://www.wikidata.org/wiki/Q21201
concepts[18].display_name	Sociology
concepts[19].id	https://openalex.org/C149923435
concepts[19].level	1
concepts[19].score	0.0
concepts[19].wikidata	https://www.wikidata.org/wiki/Q37732
concepts[19].display_name	Demography
concepts[20].id	https://openalex.org/C199360897
concepts[20].level	1
concepts[20].score	0.0
concepts[20].wikidata	https://www.wikidata.org/wiki/Q9143
concepts[20].display_name	Programming language
keywords[0].id	https://openalex.org/keywords/computer-science
keywords[0].score	0.6417620182037354
keywords[0].display_name	Computer science
keywords[1].id	https://openalex.org/keywords/quantile-regression
keywords[1].score	0.5888808369636536
keywords[1].display_name	Quantile regression
keywords[2].id	https://openalex.org/keywords/set
keywords[2].score	0.5677388310432434
keywords[2].display_name	Set (abstract data type)
keywords[3].id	https://openalex.org/keywords/coverage-probability
keywords[3].score	0.520834743976593
keywords[3].display_name	Coverage probability
keywords[4].id	https://openalex.org/keywords/trustworthiness
keywords[4].score	0.4965875744819641
keywords[4].display_name	Trustworthiness
keywords[5].id	https://openalex.org/keywords/interval
keywords[5].score	0.4946124255657196
keywords[5].display_name	Interval (graph theory)
keywords[6].id	https://openalex.org/keywords/regression
keywords[6].score	0.479521781206131
keywords[6].display_name	Regression
keywords[7].id	https://openalex.org/keywords/quantile
keywords[7].score	0.46922004222869873
keywords[7].display_name	Quantile
keywords[8].id	https://openalex.org/keywords/population
keywords[8].score	0.4299023747444153
keywords[8].display_name	Population
keywords[9].id	https://openalex.org/keywords/code
keywords[9].score	0.42871496081352234
keywords[9].display_name	Code (set theory)
keywords[10].id	https://openalex.org/keywords/machine-learning
keywords[10].score	0.38563358783721924
keywords[10].display_name	Machine learning
keywords[11].id	https://openalex.org/keywords/statistics
keywords[11].score	0.35695305466651917
keywords[11].display_name	Statistics
keywords[12].id	https://openalex.org/keywords/confidence-interval
keywords[12].score	0.34881192445755005
keywords[12].display_name	Confidence interval
keywords[13].id	https://openalex.org/keywords/artificial-intelligence
keywords[13].score	0.3415707051753998
keywords[13].display_name	Artificial intelligence
keywords[14].id	https://openalex.org/keywords/econometrics
keywords[14].score	0.32040315866470337
keywords[14].display_name	Econometrics
keywords[15].id	https://openalex.org/keywords/mathematics
keywords[15].score	0.2153456211090088
keywords[15].display_name	Mathematics
keywords[16].id	https://openalex.org/keywords/computer-security
keywords[16].score	0.07599031925201416
keywords[16].display_name	Computer security
language	en
locations[0].id	pmh:oai:arXiv.org:2311.02243
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/2311.02243
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/2311.02243
locations[1].id	doi:10.48550/arxiv.2311.02243
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.2311.02243
indexed_in	arxiv, datacite
authorships[0].author.id	https://openalex.org/A5101686970
authorships[0].author.orcid	https://orcid.org/0000-0003-2559-045X
authorships[0].author.display_name	Fangxin Wang
authorships[0].author_position	first
authorships[0].raw_author_name	Wang, Fangxin
authorships[0].is_corresponding	False
authorships[1].author.id	https://openalex.org/A5102076839
authorships[1].author.orcid
authorships[1].author.display_name	Lu Cheng
authorships[1].author_position	middle
authorships[1].raw_author_name	Cheng, Lu
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/A5070129944
authorships[3].author.orcid	https://orcid.org/0000-0002-2022-9465
authorships[3].author.display_name	Kay Liu
authorships[3].author_position	middle
authorships[3].raw_author_name	Liu, Kay
authorships[3].is_corresponding	False
authorships[4].author.id	https://openalex.org/A5036357902
authorships[4].author.orcid	https://orcid.org/0000-0002-3491-5968
authorships[4].author.display_name	Philip S. Yu
authorships[4].author_position	last
authorships[4].raw_author_name	Yu, Philip S.
authorships[4].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/2311.02243
open_access.oa_status	green
open_access.any_repository_has_fulltext	False
created_date	2025-10-10T00:00:00
display_name	Equal Opportunity of Coverage in Fair Regression
has_fulltext	False
is_retracted	False
updated_date	2025-11-06T06:51:31.235846
primary_topic.id	https://openalex.org/T10883
primary_topic.field.id	https://openalex.org/fields/33
primary_topic.field.display_name	Social Sciences
primary_topic.score	0.9821000099182129
primary_topic.domain.id	https://openalex.org/domains/2
primary_topic.domain.display_name	Social Sciences
primary_topic.subfield.id	https://openalex.org/subfields/3311
primary_topic.subfield.display_name	Safety Research
primary_topic.display_name	Ethics and Social Impacts of AI
related_works	https://openalex.org/W4206511378, https://openalex.org/W4206618949, https://openalex.org/W2526321210, https://openalex.org/W3205863630, https://openalex.org/W4318833145, https://openalex.org/W2364275385, https://openalex.org/W4388704167, https://openalex.org/W2007977664, https://openalex.org/W4376874882, https://openalex.org/W2224749288
cited_by_count	3
counts_by_year[0].year	2024
counts_by_year[0].cited_by_count	1
counts_by_year[1].year	2023
counts_by_year[1].cited_by_count	2
locations_count	2
best_oa_location.id	pmh:oai:arXiv.org:2311.02243
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/2311.02243
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/2311.02243
primary_location.id	pmh:oai:arXiv.org:2311.02243
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/2311.02243
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/2311.02243
publication_date	2023-11-03
publication_year	2023
referenced_works_count	0
abstract_inverted_index..	179
abstract_inverted_index.a	60, 99, 119, 137, 153
abstract_inverted_index.--	64, 70
abstract_inverted_index.It	134
abstract_inverted_index.ML	132
abstract_inverted_index.To	54
abstract_inverted_index.We	0, 112
abstract_inverted_index.an	22
abstract_inverted_index.at	98, 177
abstract_inverted_index.be	107, 110
abstract_inverted_index.in	49, 169
abstract_inverted_index.is	47, 174
abstract_inverted_index.it	27
abstract_inverted_index.of	18, 52, 67, 166
abstract_inverted_index.on	42, 152
abstract_inverted_index.to	9, 73, 109, 123, 140, 149
abstract_inverted_index.we	58
abstract_inverted_index.(1)	77
abstract_inverted_index.(2)	89
abstract_inverted_index.EOC	125, 151
abstract_inverted_index.Our	172
abstract_inverted_index.The	15
abstract_inverted_index.and	12, 46, 88
abstract_inverted_index.any	130
abstract_inverted_index.are	86
abstract_inverted_index.for	80, 93, 129
abstract_inverted_index.new	61
abstract_inverted_index.not	29
abstract_inverted_index.our	167
abstract_inverted_index.set	139
abstract_inverted_index.the	43, 50, 90, 94, 103, 164
abstract_inverted_index.two	75
abstract_inverted_index.(ML)	5
abstract_inverted_index.EOC,	144
abstract_inverted_index.EOC.	171
abstract_inverted_index.Fair	115
abstract_inverted_index.aims	72
abstract_inverted_index.code	173
abstract_inverted_index.does	28
abstract_inverted_index.fair	2
abstract_inverted_index.from	143
abstract_inverted_index.more	35
abstract_inverted_index.rate	92
abstract_inverted_index.set,	155
abstract_inverted_index.test	154
abstract_inverted_index.that	71
abstract_inverted_index.then	145
abstract_inverted_index.true	44
abstract_inverted_index.with	83, 126
abstract_inverted_index.work	17
abstract_inverted_index.(EOC)	69
abstract_inverted_index.Equal	65
abstract_inverted_index.bound	141
abstract_inverted_index.equal	31
abstract_inverted_index.first	135
abstract_inverted_index.label	45
abstract_inverted_index.rates	33, 79
abstract_inverted_index.study	1
abstract_inverted_index.these	56
abstract_inverted_index.under	6
abstract_inverted_index.width	128
abstract_inverted_index.(e.g.,	38
abstract_inverted_index.Binned	114
abstract_inverted_index.across	34
abstract_inverted_index.biased	48
abstract_inverted_index.close,	87
abstract_inverted_index.enable	10
abstract_inverted_index.entire	95
abstract_inverted_index.groups	37, 82
abstract_inverted_index.level.	101
abstract_inverted_index.method	122, 168
abstract_inverted_index.narrow	108
abstract_inverted_index.should	106
abstract_inverted_index.tackle	55
abstract_inverted_index.width.	160
abstract_inverted_index.(BFQR),	118
abstract_inverted_index.achieve	74
abstract_inverted_index.improve	124
abstract_inverted_index.machine	3
abstract_inverted_index.models.	133
abstract_inverted_index.notion.	25
abstract_inverted_index.propose	59, 113
abstract_inverted_index.remains	97
abstract_inverted_index.results	162
abstract_inverted_index.seminal	16
abstract_inverted_index.similar	84
abstract_inverted_index.trained	131
abstract_inverted_index.Coverage	68
abstract_inverted_index.Further,	102
abstract_inverted_index.However,	26
abstract_inverted_index.Quantile	116
abstract_inverted_index.coverage	32, 78, 91
abstract_inverted_index.fairness	24, 63
abstract_inverted_index.females)	40
abstract_inverted_index.hold-out	138
abstract_inverted_index.interval	159
abstract_inverted_index.learning	4
abstract_inverted_index.maintain	150
abstract_inverted_index.outcomes	85
abstract_inverted_index.proposed	21
abstract_inverted_index.publicly	175
abstract_inverted_index.reliable	11
abstract_inverted_index.available	176
abstract_inverted_index.conformal	147
abstract_inverted_index.deviation	142
abstract_inverted_index.different	81
abstract_inverted_index.guarantee	30
abstract_inverted_index.improving	170
abstract_inverted_index.intervals	105
abstract_inverted_index.leverages	146
abstract_inverted_index.meanwhile	156
abstract_inverted_index.Regression	117
abstract_inverted_index.assessment	51
abstract_inverted_index.calibrates	136
abstract_inverted_index.coverage''	20
abstract_inverted_index.low-income	39
abstract_inverted_index.optimizing	157
abstract_inverted_index.population	96
abstract_inverted_index.prediction	104, 148, 158
abstract_inverted_index.predictive	7
abstract_inverted_index.reasonable	127
abstract_inverted_index.Opportunity	66
abstract_inverted_index.``equalized	19
abstract_inverted_index.demonstrate	163
abstract_inverted_index.properties:	76
abstract_inverted_index.trustworthy	13
abstract_inverted_index.uncertainty	8
abstract_inverted_index.Experimental	161
abstract_inverted_index.conditioning	41
abstract_inverted_index.fine-grained	36
abstract_inverted_index.informative.	111
abstract_inverted_index.limitations,	57
abstract_inverted_index.uncertainty.	53
abstract_inverted_index.effectiveness	165
abstract_inverted_index.predetermined	100
abstract_inverted_index.post-processing	121
abstract_inverted_index.decision-making.	14
abstract_inverted_index.distribution-free	120
abstract_inverted_index.uncertainty-aware	23, 62
abstract_inverted_index.https://github.com/fangxin-wang/bfqr	178
cited_by_percentile_year
countries_distinct_count	0
institutions_distinct_count	5
sustainable_development_goals[0].id	https://metadata.un.org/sdg/1
sustainable_development_goals[0].score	0.4000000059604645
sustainable_development_goals[0].display_name	No poverty
citation_normalized_percentile